Authors′ Reply

To the Editors: We thank Dr Teofili and colleagues for their comments regarding our recent manuscript entitled “Association of HIV-Infection and Antiretroviral Therapy with Levels of Endothelial Progenitor Cells and Subclinical Atherosclerosis.”1 Endothelial progenitor cells (EPCs) are bone marrow derived cells characterized by the expression of both hematopoietic (CD34, CD133) and endothelial surface markers (such as KDR, VE-cadherin, CD31, and von Willebrand factor) and by their capacity to contribute to vascular homeostasis.2 In general, 2 approaches have been used to isolate and identify EPCs: (1) cell cultured assays from blood and (2) selection of subpopulations based on surface markers by flow cytometry. Culture assays may expand the cells, although it is unclear how cultured conditions may affect cell phenotypes. By flow cytometry, cell populations can be isolated without ex vivo manipulation, although there are no accepted surface markers that precisely identify the different subtypes of progenitor cells and standardized protocol of the technique.3,4 In the general population, an association between low levels of CD34+/KDR+ cells and increased carotid intima-media thickness has been identified.5 Consistently, previous data demonstrated that their levels independently predicted cardiovascular events in patients with different cardiovascular risk and even in healthy subjects.6 So, the aim of our study was to investigate whether this association could also occur in HIV-infected patients. We used the same methodology, flow cytometry, to make comparisons with previous studies, although acknowledging that the main limitation of our study was the lack of specific cell surface markers that precisely identify the different subtypes. In fact, most of the markers used to identify EPCs or smooth muscle progenitor cells (CD34, KDR, CD14, and α-SMA) can also be expressed by mesenchymal or hematopoietic cells.3,4 We acknowledge that many issues in the field of unequivocal molecular characterization of EPCs still remain unresolved. However, since the first identification of EPCs by Asahara et al7 in 1997, significant researches have been taken to reach a better definition and detailed functional characterization of these cells. In this sense, since the original study by Asahara et al proposing CD34 and KDR as markers of EPCs, many other articles have been published using this combination to identify EPCs. However, some studies have proposed that CD34+/KDR+ cells cannot be considered true EPCs because they do not convincingly differentiate in endothelial cells in vivo.2,3 Alternatively, endothelial colony forming cells (ECFCs), isolated by techniques based on cell cultures, have been proposed as the only cells that display all the properties of true EPCs,3,4 as has been noted by Teofili et al. Unfortunately, at present, no specific antigens have been identified that can unequivocally discriminate ECFCs from other circulating blood elements.3,4 Therefore, our study in a large population of HIV-infected patients would have not been feasible using a culture-based protocol. At present, there is a consensus that several cell types with different phenotypes may participate in revascularization process, initiating, facilitating, and regulating the incorporation of ECFCs to injured endothelium.4 So, to avoid confusion with the term EPCs, it has been proposed to strictly define all the cells involved in revascularization by their phenotypic and/or functional properties. In this sense, in our article, we have identified specific progenitor cells, and we have indicated that our conclusions should be restricted to these progenitor phenotypes. Assuming all these limitations, our data demonstrated that HIV-infected patients showed lower number of CD34+/KDR+ and CD34+/VE-cadherin+ cells than healthy subjects and that antiretroviral therapy (ART), mainly protease inhibitor-based regimen, is the most powerful predictor of their levels.1 These results confirm previous studies8,9 and extend them to other phenotypes of EPCs. On the opposite, 2 groups have reported that circulating EPCs (CD45dim/CD34+/KDR+ and CD34+/KDR+, respectively) are not reduced in ART-naive HIV-infected patients with respect to noninfected subjects.10 However, important differences between both studies and ours could account for these discrepancies. Costiniuk et al10 excluded from their study women, current smokers, and patients with established cardiovascular risk factors or receiving ART; these factors being important modifiers of EPC levels.1 More intricate, we cannot compare our data with those reported by Papasavvas et al11 because these authors used cryopreserved peripheral blood mononuclear cells. Teofili et al12 reported that circulating colony-forming unit-endothelial cells (CFU-ECs), but not ECFCs, were significantly reduced in HIV-positive patients. Interestingly, similar to CD34+/KDR+ cells, CFU-ECs cannot be considered true EPCs because of their low capacity of proliferation and endothelial differentiation.3,4 Finally, the measurement of EPCs as biomarkers, and any other, in large clinical studies requires simple, rapid, and reproducible methods as flow cytometry. Indeed, rather than being highly biologically informative, it is mandatory to demonstrate a strong statistical association with several clinical aspects of the disease. In this sense, EPCs, defined as CD34+/KDR+ cells, have been associated with the occurrence of a first major cardiovascular event in patients at different risks or even in healthy subjects.4–6 Moreover, restoration of EPC number and/or their functionality is possible through current therapies for cardiovascular risk factors,6 suggesting that they could also be a promising tools for measuring therapeutic efficacy. However, culture-based studies will allow continued understanding of EPC biology for using them as cellular regenerative therapies.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.