Prostate Cancer Stem Cell Therapy: Hype or Hope?
Posted: 12/02/2008; Prostate Cancer Prostatic Dis. 2008;11(4):316-319. © 2008 Nature Publishing Group
Abstract and Introduction
The stem cell concept of cancer suggests that each cancer contains a small fraction of stem cells responsible for the maintenance and progression of the disease. The implication of this concept is that by targeting and killing the cancer stem cells, it may be possible to improve survival or even cure the disease. Prostate cancer stem cell therapy is a valid goal to aim for, but there are massive hurdles to overcome, even if the concept is shown to be correct.
The cancer stem cell concept suggests that there is a small population of stem cells that is responsible for the growth and progression of cancer. It is hoped that targeting and killing this small fraction of the cancer cells might cure the tumour, or, at least, improve survival. This review summarizes the evidence, and considers whether prostate cancer patients will ultimately benefit from cancer stem cell therapy.
Cancer therapy has been based on the notion that most of the cancer cells have a similar capacity to divide, metastasize and kill the host. Consequently, current therapies, including surgery, radiotherapy and chemotherapy, aim to eradicate or kill every cancer cell. The cancer stem cell concept suggests a fundamentally different approach. As only a small fraction of the cancer cells are responsible for the growth and progression of the disease, it is this small population of cancer stem cells that must be eradicated to cure the cancer. The goal now is to translate preliminary data on the biology of normal and cancer stem cells into some form of cancer therapy””a daunting long-term task of uncertain outcome. While cure might be an unrealistic hype for the majority of cancers, there is a reasonable hope that if such therapy could be developed over the next 10-20 years, it may increase survival for at least some men with prostate cancer.
The cancer stem cell concept is not new. The pioneering work of Ernest McCulloch and James Till in Toronto in the 1950s and 1960s received one of science’s highest accolades, a Lasker award, in 2005. The degree of interest in cancer stem cells since the 1950s has fluctuated. The most recent wave of interest results from two technical developments. The first of these developments is the identification of cell surface antigens on the surface of stem cells by which stem cells can be sorted and isolated (using FACS; fluorescence-activated cell sorting). The second development was the breeding and supply of immunosuppressed mice in which human stem cells can be grown to their full developmental potential. These developments have been utilized by John Dick to re-energize the cancer stem cell concept, two generations of Toronto scientists on from McCulloch and Till.
What is the evidence for cancer stem cells? Much of the early interest in cancer stem cells was associated with the concept that cancer is a clonal disease, derived from a single cell. A clonal origin for cancer fits well with the concept that cancer is a stem cell disease derived from the progeny of the first genetically altered cell. The idea that only a tiny fraction of the cancer cells have indefinite self-renewal capacity (that is, are stem cells) was reinforced by ‘stem cell assays’, using which human cancer cells were grown in suspension culture in the laboratory.[2,3] Only a tiny fraction of the cells, usually less than 1%, are capable of generating colonies, a property equated with self-renewal capacity, despite colony-forming ability not being exclusive to stem cells. Clonogenic assays were used 50 years ago by McCulloch and Till both in vitro (in petri dishes) and in vivo (in mice, for example, spleen colony-forming assay) and are still used today as a means of characterizing stem cells.[4,5]
Ten years ago, most cancer scientists dismissed the cancer stem cell concept, arguing that most if not all cancer cells have the potential for indefinite self-renewal. Two papers from John Dick’s laboratory were crucial in changing this attitude.[6,7] These papers demonstrated that AML (acute myeloid leukaemia) contains a stem cell fraction having the same cell surface characteristics as normal haematopoietic stem cells, and it is only these cells, which are capable of regenerating AML in immunosuppressed animals. In an elegant series of studies, Dick and colleagues went on to show that AML recapitulates patterns of differentiation similar to those seen in normal haematopoietic cells, again consistent with stem cell theory.
Solid tumours such as prostate cancer pose greater technical challenges than leukaemias. It is more difficult to produce a single-cell suspension from a solid cancer and isolate and grow the stem cells both in vitro and in vivo. Nevertheless, the findings in AML were soon replicated in breast cancer. Biopsies, mainly ascites, were FACS-sorted on the basis of cell surface markers thought to enrich for stem cells and transplanted to the mammary fat pads of NOD-SCID (non-obese diabetic, severe combined immunodeficient) mice. These putative cancer stem cells had at least a 1000-fold greater capacity for regenerating tumours mice than the cell population as a whole.
Similar results using human cancer biopsies have been obtained for brain cancers,[10,11,12,13] colon[14,15] and head and neck cancer. In all these studies, the stem cell-enriched fraction of cells was transplanted to immunosuppressed mice to generate tumours that recapitulated the original histopathology. However, in every case, only a fraction of the stem cell-enriched fraction could regenerate cancers, and, at least, 100 sorted cells were required to produce tumours. The ultimate evidence for a human cancer stem cell would be the identification of a single cell selected from a human cancer that regenerated the tumour in an experimental animal.
A secondary question is at which stage of cell differentiation does the cancer stem cell arise? This question gives rise to many sterile semantic arguments. It is possible that cancer arises in a cell at any stage of differentiation, from the most primitive stem cell to the most differentiated tissue-specific cell. But the early events are most likely to occur in normal stem cells, as it is only these cells, which live long enough to accumulate the several genetic changes required for an invasive cancer to develop. However, once one or more initiating genetic changes have occurred in the cancer stem cell, all the downstream cells will also contain this change(s), in which case, it is possible that one of the downstream cells acquires not only the properties of a stem cell, but also the additional genetic changes that allow the cancer to progress to the next step in its development. Evidence for such a serial progression or clonal evolution of cancer stem cells has been demonstrated for CLL (chronic lymphocytic leukaemia).
There is one study that suggests that cancer can also arise from circulating stem cells that are more primitive than the tissue-specific adult stem cell. Chronic infection of mice with Helicobacter resulted in cells from the bone marrow populating the stomach and developing cancer. Despite the plausibility of the data and its compelling rationale, this study has yet to be replicated.
What of the prostate? It is lagging behind most other cell types as usual. Evidence has been published for normal prostate epithelial stem or progenitor cells in the human[18,19,20] and mouse prostate.[21,22,23,24,25] Mouse data suggest that the stem cell compartment lies towards the base of the ducts near where they join the urethra.[26,27,28] Prostate tissue can be reconstituted by embryonic stem cells.
Evidence for prostate cancer stem cells is limited. One study found evidence for prostate cancer stem cells using immortalized cell lines. Another study grew cells from small number of radical prostatectomy specimens. However, there was no direct confirmation that the cells that grew were cancer cells, and it was suggested by Litvinov et al.  that there is a strong possibility that the cells described by Collins et al.  are of normal rather then cancer origin. When cell cultures were derived from radical prostatectomy samples, normal cells grew rather than cancer cells in medium containing low levels of calcium. The specification of culture conditions that favour the growth of prostate cancer cells over normal prostate epithelial cells may be required to reproducibly isolate human prostate cancer stem cells.
Research on the isolation and characterization of cancer stem cells was in vogue 25-30 years ago due to the popularity of the ‘human tumour stem cell assay’. Large numbers of publications purported to clone human cancer cells in suspension culture, and many attempts were made to use the assay to individualize treatment. But many of these studies lacked scientific rigour and most of the so-called human tumour stem cell colonies were technical artefacts, due to a variety of reasons including the growth of normal cells. Today, we may be repeating these errors. Consequently, it is imperative that any study today uses serial single-cell cloning and demonstrates that the colonies originate from cancer and not normal cells. Evidence for a cancer origin can be obtained relatively easily by allelotyping or SNP analysis to demonstrate that the genetic changes in the cancer are recapitulated in the human cancer cell colonies. If possible, the cells should be serially propagated as cancers in immunosuppressed animals, as described for other cancers and leukaemias.
Cancer stem cell culture is demanding, relatively unproductive and laborious in terms of generating experimental data. The lure is the potential impact on the survival of cancer patients. If cell surface markers can be identified that are specific for cancer stem cells, which can then be used as targets to kill the cells, then there is the possibility of highly selective and specific cancer stem cell therapy. Such treatment has the potential to target and kill metastatic disease and increase the survival of cancer patients.
However, some cynical armchair scientists might suggest that the major insight provided by the cancer stem cell hypothesis is an explanation of why cancer is so hard to cure. It seems probable that the cancer stem cells are responsible for disease progression, are relatively resistant to chemotherapy and radiotherapy and are the cells responsible for metastasis. It seems much easier to kill downstream cells than it is to kill the cancer stem cells. This may explain why the vast majority of metastatic cancers respond for relatively short periods to chemotherapy and hormone therapy before the stem cells recover and resume their inexorable growth to kill their hosts. Most men with prostate cancer benefit in the short term from hormone therapy, but none are cured””presumably prostate cancer relapse following hormone therapy is associated with the growth of the cancer stem cells.
The pessimist will see the reasons why cancer stem cell therapy might not work, or that even if it does achieve some responses, it has as little impact as 99% or more of the other novel treatments tested in the last 30 years. First and foremost is the bane of cancer therapy””the lack of or limited selectivity against cancer cells. There needs to be a specific target on the stem cell or in the stem cell niche that is not duplicated on cells with some other vital function. Furthermore, for most cancers, the target may need to be tissue specific to avoid hitting stem cells in other vital organs.
Even if the problem of selectivity is overcome, the stem cell fraction in a cancer is not a sitting target. The cancer may start out in a normal stem cell, but all the progeny of the cancer stem cell will also contain all the genetic changes acquired by the initiating cell. In consequence, it is possible that some downstream cells will acquire stem cell properties and become responsible for the progression of the cancer, leaving the initiating cells behind, and, perhaps irrelevant, as it appears to happen in CML. Whether these cells will have the same cell surface properties or markers as the initiating stem cells is questionable. Furthermore, the stem cell seems capable of adapting to current treatments with facility.
What of the search for specific markers, probably on the cell surface, using which cancer stem cells can be targeted? Such studies are ongoing, but a universal target has yet to emerge, if indeed one exists. Prostate and breast cancers have the advantage over leukaemias and brain tumours that the corresponding normal stem cell compartment is not vital and can be dispensed with. Consequently, a tissue-specific stem cell therapy is a goal worth pursuing for prostate cancer, although it still has a very long way to go. Nevertheless, there has been no real improvement in the survival of men with metastatic prostate cancer since castration was introduced by Huggins in the 1940s. So, it is comforting for the next generation of men to live in the hope of an effective prostate cancer stem cell therapy, even though with more awareness of the limitations and less hype.