| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Innovations and Challenges in Renal Cancer |
Authors' Affiliation: Divisions of Hematology-Oncology, Nephrology, Interdisciplinary Medicine, and Biotechnology, Department of Medicine, Beth Israel Deaconess Medical Center, Boston, Massachusetts
Requests for reprints: Rupal S. Bhatt, Beth Israel Deaconess Medical Center, RW563, 330 Brookline Avenue, Boston, MA 02215. E-mail: rbhatt{at}bidmc.harvard.edu.
 |
Abstract |
|---|
 Top Abstract Potential Use of Surrogate...Noninvasive MarkersFuture PerspectivesOpen DiscussionReferences |
|---|
|
Potential Use of Surrogate Biomarkers |
|---|
|
TopAbstractPotential Use of Surrogate...Noninvasive MarkersFuture PerspectivesOpen DiscussionReferences |
|---|
Despite showing a high proportion of tumor shrinkage, antiangiogenic therapy produces few, if any, complete tumor regressions and patient tumors typically develop resistance after a median of 6 to 12 months of receiving therapy. Thus, biomarkers of resistance or relapse that offer mechanistic insight are just as important to identify as markers of response. Although little is known about the mechanisms of resistance to antiangiogenic therapy, it has been postulated that resistance develops as a result of activation of compensatory pathways driving angiogenesis in the setting of VEGF receptor-2 blockade. Resistance could develop by three possible mechanisms: (a) incomplete inactivation of the VEGF pathway; (b) activation of other hypoxia-inducible factor–driven genes, such as CXCR4 and TGF-
; and (c) selection of tumor cell populations able to survive in the presence of VEGF receptor-2 blockade, perhaps by activation of VEGF-independent pathways such as fibroblast growth factor and interleukin-8 (4, 5). It should also be noted that neovascularization encompasses more that just angiogenesis. Thus, tumors may potentially develop resistance by other less well-studied mechanisms such as vessel co-option, vasculogenesis (e.g., stromal cell–derived factor 1), and vascular mimicry.
To develop useful biomarkers, one must consider the characteristics of a good biomarker. These include cost-effectiveness, low baseline levels in normal individuals, accessibility by noninvasive means such as blood and urine, robustness in the clinical setting, and reproducibility in multiple clinical centers. Although noninvasive means of following up patients is the ultimate goal, we will likely need to be guided initially by information obtained by invasive methods. For example, there is much interest in obtaining tissue biopsy specimens from patients who are receiving targeted therapies to assess the molecular features of the tumor exposed to a specific agent. The current hope is that pathologic features of a tumor will have clinical and laboratory correlates. This paradigm is addressed in the neoadjuvant administration of targeted agents with biopsy specimens obtained before therapy and then with tissue acquired at the time of nephrectomy. Similarly, surgery or radiofrequency ablation of limited metastatic disease could be a valuable source of tumor tissue that has been exposed to antiangiogenic therapy. It is hoped that the findings in tissue will guide the selection of biomarkers in patient blood. Correlation of pathologic data with markers in patient blood will likely lead to useful surrogate markers of the local tumor angiogenic state.
One recognized caveat of pathologic analyses is sampling error. Indeed, tumors may have differential areas of response and/or relapse to therapy within them. Pathologic correlation with imaging to select appropriate areas of a tumor for study may prove a useful way of minimizing sampling bias.
|
Noninvasive Markers |
|---|
|
TopAbstractPotential Use of Surrogate...Noninvasive MarkersFuture PerspectivesOpen DiscussionReferences |
|---|
Circulating cytokines. Classes of proteins that may prove useful to measure are those that promote or suppress angiogenesis, including cytokines/growth factors such as VEGF and its family members, basic fibroblast growth factor, and several chemokines (both proangiogenic and antiangiogenic members), although no single predictive cytokine has been found to date. Soluble shed receptors and the proteases that cleave them, such as soluble Flt-1, soluble VEGF receptor 2, soluble vascular cell adhesion molecule, and various matrix metalloproteinases, could also prove to be useful biomarkers. The use of cytokines as measures of angiogenic activity can be complicated by the fact that platelets contain and could release many angiogenic and antiangiogenic factors that could confound accurate measurements in patient samples.
In vitro functional assays. Another potential means of measuring the angiogenic state of patient plasma is by in vitro biological assays in which cultured endothelial cells are incubated with patient plasma and then subjected to various analyses, including proliferation, migration, and endothelial tube formation (6). Although these assays have the benefit of providing an integrated assessment of the cumulative angiogenic capability of a patient's plasma, they are labor-intensive, expensive, and difficult to standardize for large multicenter trials.
Circulating cells. A promising area of antiangiogenic monitoring is the measurement of circulating endothelial cells (CEC) in the peripheral blood of patients. This may be achieved by using cell-surface marker–based purification strategies, in vitro colony formation, or reverse transcription-PCR. Reverse transcription-PCR for markers such as CD146 and VE-cadherin has been shown to correlate with CEC number (7, 8). However, very few data about measurement of RNA in patient peripheral blood samples are available; thus, this is a method that still needs considerable further development. Moreover, the optimal collection tube and processing method for RNA purification is still being optimized.
Currently, cell-surface marker–based strategies are the most commonly used means of quantitating CECs. It is thought that, under physiologic conditions, blood vessels shed endothelial cells into the circulation and endothelial precursor cells are mobilized from the bone marrow, which are likely to participate in the formation of new vessels. This process can be perturbed by proangiogenic neoplasms and by antiangiogenic therapies. The total CECs are composed of mature CECs and circulating endothelial progenitors (also referred to as endothelial precursor cells). Multiple surface markers have been used to define CEC populations by multiparameter flow cytometry. Strategies make use of endothelial markers such as CD31 and CD146, the absence of leukocyte markers such as CD45, and the absence of the stem cell marker CD133 to define mature CECs, with circulating endothelial progenitors having the same characteristics as mature CECs with the exception that they are CD133+. In turn, each of these populations could be viable or apoptotic.
Several studies have shown that with antiangiogenic therapy, the number of these cells changes. For example, Beaudry et al. (9) showed in a murine Lewis lung carcinoma model that in tumor-bearing mice subjected to VEGF inhibition, the number of mature CECs increased. Beerepoot et al. (10) found that patients with progressing cancer had higher levels of viable mature CECs than healthy patients or patients with stable disease. Mansuco et al. (11) found that, in breast cancer patients treated with metronomic dosing of chemotherapy, patients who showed clinical benefit had an increase in apoptotic CECs from baseline to the second month of therapy. Ebbinghouse et al. (12) showed that baseline CEC numbers may be a possible prognostic marker for patients treated with the thrombospondin mimetic peptide ABT-510. A hypothetical model for CEC kinetics that is consistent with the above data is shown in Fig. 1 . However, a recent study found that treatment with an antivascular agent led to an increase in circulating endothelial progenitors early in the course of therapy (13). These studies have highlighted the fact that sampling at multiple time points may be critical for the interpretation of data from clinical trials.
|
Although CECs are the most commonly studied circulating cells, there is a possibility that other minor populations of circulating cells, some of bone marrow origin, will serve as useful biomarkers, although this remains speculative at the moment (15–18).
Imaging. Monitoring of antiangiogenic therapy could be achieved by imaging the tumor vessels themselves or the effect of such therapy on the tumor cells. [18F]Fluorodeoxyglucose positron emission tomography, for example, visualizes metabolic changes in tumor cells, likely as a result of tumor cell hypoxia induced by angiogenic blockade. Currently, vessel anatomy and flow are being assessed by novel imaging modalities such as diffusion contrast-enhanced magnetic resonance imaging, arterial spin-labeling magnetic resonance imaging, and computed tomographic perfusion. Molecular imaging of tumor vessel-specific markers or of endothelial apoptosis could be developed in the near future, and these techniques could provide extremely useful surrogates of angiogenic activity and have the benefit of providing a view of the tumor itself and not of a surrogate for tumor activity (i.e., the patient blood). These imaging modalities have the added benefit of viewing intratumoral heterogeneity in response to targeted agents. A more comprehensive view of imaging modalities as a means of following up patients is described by Rosen and Schnall (19).
|
Future Perspectives |
|---|
|
TopAbstractPotential Use of Surrogate...Noninvasive MarkersFuture PerspectivesOpen DiscussionReferences |
|---|
The future of this field also involves the collection of a large number of patient samples on antiangiogenic therapy. This involves collection of blood for analysis of plasma and peripheral blood mononuclear cells and for nucleic acid preparation. Ideally, the collection of samples would include serial samples of patients before therapy, while receiving therapy, and at the time of disease progression. However, the analysis of serial patient samples needs to be done with care. For example, there may be multiple mechanisms for relapse. CECs may serve as a common early indicator of relapse, with different patients showing different compensatory pathways to overcoming VEGF blockade. Thus, care should be taken to consider individual differences as this field progresses. This should lead to optimization and individualization of therapy for patients, especially because multiple choices for antiangiogenic therapy exist and more are being developed.
In addition to development of a large database of patient blood samples, it is imperative that tissue samples from patients before therapy and while receiving therapy be collected for pathologic analyses to best guide the study of noninvasive biomarkers. Correlation of noninvasive and invasive analytic methods will move this field forward toward its goal of optimizing the timing, dose, and proper selection of initial and salvage therapies and to selection of appropriate patients for these therapies.
|
Open Discussion |
|---|
|
TopAbstractPotential Use of Surrogate...Noninvasive MarkersFuture PerspectivesOpen DiscussionReferences |
|---|
Dr. Sukhatme: There are no data in any animal models of renal cell carcinoma, but we could get some guidance from animal models of other tumor types treated with the inhibitors in question. We can also learn from correlative studies in trials outside renal cell carcinoma with similar drugs, such as the breast cancer and colon cancer trials. I am also in favor of trying to do intensive studies on small groups, as pilots first, to generate hypotheses.
Dr. Figlin: There are some data that you should be aware of, although they have not yet appeared in the abstract form. We published at American Society of Clinical Oncology last year a report of more than 100 kidney cancer patients who were evaluated on an ABT-510 thrombospondin mimetic trial looking at CECs. One noted that the frequency of CECs is different by Memorial Sloan-Kettering Cancer Center risk group. Therefore, one can look at these cells prognostically as a surrogate for clinical parameters of risk. Then, if you are going to look at the same patient over time, you may be able to see bigger changes that might correlate with treatment.
Dr. George: Maybe another way of phrasing this would be to form it around a hypothesis. There is enough preclinical modeling in basic science behind this to form those hypotheses and then test. We will not form anything definitive in small single-arm studies, but we can't get to the large randomized populations unless we have at least tested some of these hypotheses preliminarily in some of these small settings.
Dr. McDermott: As we create a state of more complete VEGF deprivation with combination therapy and more potent VEGF receptor inhibitors, do you think we are going to see more cardiovascular toxicity in our patients?
Dr. Sukhatme: I do. Some of the effects occur because of the effects of VEGF deprivation on endothelial cells, which we know a fair amount about. These effects include hypertension, proteinuria, posterior leukoencephalopathy, and possibly thrombosis. In addition, direct effects on cardiac function may occur since VEGF receptors are expressed on cardiac myocytes and such effects may only manifest themselves clinically over longer periods of time.
Dr. McDermott: In the three kidney cancer patients we treated on a very potent inhibitor of the VEGF receptor, one had two transient ischemic attacks and stopped taking the drug because of high blood pressure. The second patient came in after taking the drug for a year and a half with fatigue and small pleural effusions. We decided to perform an echocardiogram. His ejection fraction was 15%, so we took him off the drug. His ejection fraction 6 months later increased to 50%. The third patient, who had been tolerating the drug for a year and a half, just recently had a non–Q-wave myocardial infarction. While these events may not be drug related, I tend to think that they reflect effective, long-tern inhibition of the VEGF receptor pathway.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Presented at the Second Cambridge Conference on Innovations and Challenges in Renal Cancer, March 24-25, 2006, Cambridge, Massachusetts.
Received 8/ 2/06; revised 11/ 6/06; accepted 11/10/06.
|
References |
|---|
|
TopAbstractPotential Use of Surrogate...Noninvasive MarkersFuture PerspectivesOpen DiscussionReferences |
|---|
-deficient colon cancer cells. Nat Med 2005;11:992–7.[Medline]This article has been cited by other articles:
|
|
 |
|
  S. Senan and E. F. Smit Design of Clinical Trials of Radiation Combined with Antiangiogenic Therapy Oncologist, April 1, 2007; 12(4): 465 - 477. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
