Tel: +44 203 311 8295, Fax: +44 208 846 1433
The anti-tumor effects of human immunodeficiency virus protease inhibitors: Ready for real time?
Article first published online: 27 OCT 2010
Copyright © 2010 UICC
International Journal of Cancer
Volume 128, Issue 1, pages 1–2, 1 January 2011
How to Cite
Stebbing, J. and Bower, M. (2011), The anti-tumor effects of human immunodeficiency virus protease inhibitors: Ready for real time?. Int. J. Cancer, 128: 1–2. doi: 10.1002/ijc.25590
- Issue published online: 27 OCT 2010
- Article first published online: 27 OCT 2010
- protease inhibitors;
- matrix metalloproteinases;
- Kaposi sarcoma;
In this issue of the International Journal of Cancer, Barbara Ensoli's team has further developed the proposition that human immunodeficiency virus (HIV-1) protease inhibitors or their analogues, may be a useful new class of anti-cancer agents, capable of targeting both matrix metalloproteinases and the proteasome. The history of this can be dated back to the findings that increasing use of antiretroviral therapy has led to a decline in the incidence of Kaposi sarcoma1, 2 and individual lesions themselves regress during treatment,3 especially with use of protease inhibitors.4
Despite these observations, there has been a lack of consensus with regards to the mechanism of this decline in Kaposi sarcoma incidence and the mechanism of regression. The most obvious suggestion has been that antiretrovirals lead to immune reconstitution. A reduction in immunosuppressive therapy prescribed to prevent organ transplant rejection can lead to regression of Kaposi sarcoma.5, 6 Here and in the setting of HIV, it is well known that adaptive, cellular immune responses are a critical defence against chronic viral infections and cytotoxic T lymphocyte (CTL) activity (absolute levels of which appear prognostic7) has been detected against Kaposi sarcoma-associated herepesvirus lytic and latent proteins.8 The same probably applies to certain arms of the innate immune system, especially those concerned with natural killer cells.9 We have also shown flare of Kaposi sarcoma during immune reconstitution.10
Next, antiretrovirals may reduce levels of HIV-1 Tat (transactivating protein), known for example to be angiogenic in transgenic mice.11 A third hypothesis has centred around potential direct anti-cytokine and anti-angiogenic effects of antiretrovirals, separate to the aforementioned issues connected to Tat. Supporting this, Barbara Ensoli's team have previously shown that systemic administration of the protease inhibitors saquinavir and indinavir to nude mice blocked the development of, and induced regression of, angioproliferative sarcoma-like lesions promoted by primary human Kaposi sarcoma cells, basic fibroblast growth factor (bFGF), or bFGF and vascular endothelial growth factor (VEGF) combined.12 These drugs were also found to block bFGF or VEGF-induced angiogenesis in a chorioallantoic membrane assay with a potency similar to paclitaxel, also used as an effective treatment here.13 A final theory has concerned itself with a direct action of protease inhibitors on Kaposi sarcoma-associated herpesvirus itself, although supporting evidence here seems speculative.
Thus far, a general problem with some of these ‘theories’ is the observation that Kaposi sarcoma is well known in HIV-negative individuals, and non-protease inhibitor antiretrovirals may also lead to resolution of lesions,14 even as monotherapy.15, 16
To take this translational research further, the paper by Barbara Ensoli's team investigated the effects of indinavir or saquinavir on the growth of aggressive and prevalent human tumors in nude mice exposed to cell lines of lung, breast, colon and hepatic carcinomas origin. In contrast to much of the data above, the animal model was free of viruses and T cells, thus rendering potential effects on immune reconstitution, a direct antiviral effect on HIV or Kaposi sarcoma-associated herpesvirus, entirely obsolete. Tumors, introduced via gastric lavage of cell lines with high metastatic potential, were smaller in animals treated with protease inhibitors. However, levels of the well known neoplastic proliferation marker Ki-67 in the xenografts were higher after treatment, and while this is now an accepted endpoint to demonstrate the biologic activity of certain chemotherapeutic regimens,15, 17 this may only apply to breast cancer in clinical reality. While the authors are appropriately self-critical here suggesting that proliferation and a reduction in intratumoral microvessel density (which they consistently demonstrate) may be regulated by different mechanisms, their data do in fact show a decrease in Ki-67 levels with indinavir in their breast cancer model; of interest, therapeutic levels were generally used. In situ zymographic analyses demonstrated a significant reduction of matrix metalloproteinase activity and rather than acting on the cell cycle, the protease inhibitors may block angioinvasion. Interestingly, their unpublished observations indicate that an integrin receptor implicated in MMP-2 production may be affected.
The evidence to date indicates a direct effect of HIV protease inhibitors on fundamental proliferative cellular processes. This is not surprising considering that over the course of cellular evolution, in an effort to digest food, proteases are thought to be amongst some of the first enzymes utilised by higher organisms. The authors of this paper are now starting a phase 2 clinical trial of indanavir plus chemotherapy in HIV-1 negative advanced Kaposi sarcoma. The authors of this editorial are tempted to suggest that studies in other cancers are warranted, in view of the strength of the data on this subject, confirmed by in vivo models here. The limited toxicity of these drugs in reality, developed with remarkable speed and precision since 1983 when HIV was first described (compared to the slow progress in oncology18), is a further relevant consideration here.