|
|||||||||
|
|
|
|
|
|
|
|
|
|
|
Blood, 15 June 2004, Vol. 103, No. 12, pp. 4381-4382.
Generating CTL by HLA transplantT-cell immunotherapy of human cancers is in most cases dependent on circumventing the tolerance of T cells to tumor-associated antigens. One approach to circumvent T-cell tolerance would be formed by HLA-mismatched bone marrow transplantation (BMT) coupled to donor lymphocyte infusion. Unfortunately, this approach is complicated by the fact that many of the infused cytotoxic T lymphocytes (CTLs) will recognize healthy tissues, causing severe graft-versus-host disease.1 In this issue, Savage and colleagues (page 4613) describe an elegant and straightforward strategy to selectively generate tumor-specific CTLs from HLA-mismatched donors. Infusion of such cells holds promise for the induction of a tumor-selective T-cell attack upon HLA-mismatched BMT.
Many of the tumor antigens that can form targets for CTL attack are self-antigens, and the leukemia-associated Wilms tumor antigen (WT1) and melanoma-associated Melan-A antigens that are used by the authors are no exception.2 If CTLs specific for these antigens can be found in patients, their affinity, and hence their activity, is often low because the high-affinity self-specific T cells have been deleted in the thymus or rendered inactive. The T cells of an HLA-mismatched individual could be a solution to this problem, as they have not been exposed to the donor's HLA and are thus not affected. However, if this heterogeneous population of T cells is transferred into the donor, the allogeneic T cells will recognize a broad range of allo-HLA/peptide combinations, leading to a general anti-allo response.
The crucial element of the technique developed by Savage and colleagues is to selectively expose T cells of the donor to the allogeneic HLA complex that contains the tumor antigen, by conjugating these HLA-peptide complexes to donor-derived B cells. In vitro stimulation of the donor T cells with the "HLA-transplanted" B cells results in an allo-HLA-restricted response that is focused on a single tumor antigen. The resulting CTLs appear to be of high affinity, consistent with the notion that they were derived from an individual in whom tolerance was not an issue. More importantly, the CTLs also seem to be selective for this HLA-peptide complex and appear to ignore the large repertoire of other HLA-peptide combinations. This is all the more remarkable in view of the fact that the approach used to generate these cells does not actively select against cross-reactivity with other allo-HLA/peptide combinations. Hence, a more extensive analysis of this issue before clinical application seems prudent. As the authors state, these lymphocytes can be given following HLA-mismatched transplantation in leukemia patients to reduce chances of relapse. In addition, a combination of an HLA-mismatched BMT plus alloreactive tumor-specific CTLs might be worth pursuing for tumors like melanoma, for which conventional BMT has not proven to be beneficial. Alternatively, this technology may be used to generate a collection of high-affinity tumor-specific T-cell receptors (TCRs). In the future, such TCRs may conceivably be used to endow autologous T cells with a new tumor specificity.3 Regardless of how these cells are used, the bottom line of these studies appears to be that the successful generation of tumor-specific CTLs can start with transplanting HLA molecules.
References
  CiteULike  Connotea  Del.icio.us  Digg  Reddit  Technorati What's this?
Related Article in Blood Online:
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|||||||
 |
 |
 |
 |
 |
 |
 |
 |
||
| Copyright © 2004 by American Society of Hematology Online ISSN: 1528-0020 | |||||||||
