Is thalidomide a true anti-angiogenic molecule in multiple myeloma?

In the March issue of this journal Patrizia Tosi and Michele Cavo reported about the use of thalidomide in multiple myeloma (MM).1 They underline that the mechanisms of action of thalidomide are poorly defined yet. Among them (some of which are only hypothesized2), one of the most controversial is whether thalidomide acts as an anti-angiogenic. The use of thalidomide in patients with endstage refractory MM proposed by Singhal et al.,3 was based on our observations,4-6 showing an increased bone marrow microvascularity in MM, and those7-9 showing that thalidomide is apoptogenic for neovasculature and inhibits angiogenesis in several experimental models. In 1994, D’Amato et al.7 demonstrated that thalidomide inhibits fibroblast growth factor-2 (FGF-2)-induced angiogenesis in a rabbit cornea micropocket assay. Later, he reported that thalidomide also inhibits vascular endothelial growth factor (VEGF) in a murine model of corneal vascularization,8 and others demonstrated that thalidomide inhibits microvessel formation in a rat aorta ring assay.9 According to its anti-angiogenic power, thalidomide has been used in solid tumors, such as recurrent glioma, breast cancer, melanoma, renal and ovarian cancer and hormone-refractory prostate cancer, producing, however, limited therapeutic activity.10-13 In contrast, its efficacy in acquired immunodeficiency syndrome (AIDS)-related Kaposi’s sarcoma is striking.14 As far as concerns MM treatment, no close relationship between microvessel density and clinical response was found in the study by Singhal et al.:3 there is no difference in the bone marrow microvascular density between responsive and resistant patients. However, the lack of significant lowering in bone marrow microvascular density after thalidomide therapy, does not fully rule out the antiangiogenic mechanism of action of the drug. Now, Cheng et al.15 suggest a relationship between bone marrow microvascularity and clinical response to thalidomide, since they demonstrated that an increased pretreatment microvascular density could predict the response. Moreover, while Rajkumar et al.16 demonstrated that VEGF secretion by MM cell lines is not influenced by in vitro treatment with thalidomide, Weber et al.17 showed that VEGF serum levels are high in responding patients. While angiogenesis inhibition remains an attractive potential mechanism of action, thalidomide has a number of other properties that could explain its activity in MM. These include disturbance of adhesion molecule expression that prevents myeloma cell interaction with bone marrow microenvironment,18 modulation of tumor necrosis factor-α production by stimulated monocytes, macrophages and neutrophils,19 and enhancement of cell-mediated immunity by direct co-stimulation of T-cells.20 To sum up, because a causal relationship between the anti-angiogenic effect of thalidomide and clinical activity against neoplastic diseases has still to be demonstrated, it is conceivable to hypothesize that multiple pathways may be involved in thalidomide acitivity in MM. Moreover, the rapid response usually observed is consistent with either a direct cytotoxic effect or an immunomodulatory effect. Further studies are warranted to establish possible correlations between elevated blood levels of VEGF and FGF-2, bone marrow angiogenesis and the likelihood of response.