Wake Forest University Baptist - Medical Center

Spontaneous Regression of Advanced Cancer in Mice

Summary Part III: Cancer in Normal Mice Can Be Cured by Treatment with Cells from Cancer-Resistant Mice

Using a previously described mouse model of cancer resistance, scientists in the Comprehensive Cancer Center have described new findings in Proceedings of the  National Academy of Science USA in which they demonstrate the ability to cure cancer in normal mice by transferring purified immune cells (white blood cells) from cancer resistant mice.  These studies show that specific types of innate immune cells, such as macrophages, can migrate to the site of cancer in a normal mouse and selectively kill all of the cancer cells without harming normal cells.  Such studies suggest that this type of mechanism might one day be used to help design a new strategy for cancer therapy in humans.

Here is an explanation of the research prepared by the scientists:

Subsequent to our original publication in 2003, genetic cancer resistance has been propagated into four more strains of mice and shown to work against a wide variety of cancer types.  While cancers injected into these unique mice are rejected, several questions needed answering about how this worked.  For example, while the mutant mice rejected tumors, it was not clear if this was an event mediated by specific cells in the mice, or if this mechanism could somehow be transferred to normal mice as a treatment for cancer at distant sites.  Findings presented in our second publication in Proceedings of the National Academy of Sciences USA in 2006 demonstrate the results of experiments to address these questions, and provide more in-depth information about how these unusual mice avoid cancer.

Experimental Cancers in Mice

 There are several different strategies for studying cancer in mice, and they vary in their ability to predict the behavior of cancers or therapies in human patients.  For example, one can divide the types of mouse cancer models into five major categories: 1) inducing endogenous (naturally occurring) cancer in mice using chemical carcinogens, 2) inducing endogenous cancer in mice using genetic manipulation, 3) allowing mice to grow old and get spontaneous cancers, 4) transplanting human cancers into immune-deficient mice that cannot reject cells from a different species and 5) transplanting aggressive cancers from other mice.  This last group represents the type of model we used in our experiments.  Some therapies that work for the first four categories have difficulty treating these more aggressive mouse cancers.  The SR/CR cancer resistant mouse was originally identified using these aggressive mouse cancers and, therefore, had to be a very effective resistance mechanism to have been detected.  One might predict that such an effective mechanism might not only be able to kill cancer cells in the original mutant mouse, but perhaps one could transfer this mechanism to normal mice, as well.

If SR/CR Immune Cells Kill Cancer, Will They Work In a Normal Mouse?

Even though our in vitro (test tube) experiments suggested that the innate immune cells themselves were responsible for tumor killing, it was still possible that this killing might only work if the rest of the mouse also expressed the same mutation.  Therefore, we placed cancer cells and immune cells from spontaneous remission/cancer resistant (SR/CR) mice together in a normal mouse to determine whether the cancer cells could survive.  Without the SR/CR immune cells, such cancers grow rapidly in normal mice and the mice die in 3-4 weeks.  But, when these cancer cells were injected together with the SR/CR immune cells, the tumor was killed.  Thus, the environment in a normal mouse still allowed these cells to work.  This suggests that no other cell type or soluble factor in the mice is required to allow the immune cells to function and kill cancer.

A more difficult experiment was to inject a normal mouse with cancer cells and allow the tumor to implant and grow, then inject the SR/CR immune cells at a later time.  Again, the mutant immune cells killed the cancer cells selectively, without harming the normal mouse.  Finally, we performed the most difficult challenge (shown in Figure 5), which was to inject the normal mice with cancer cells at one site (e.g., subcutaneously on the back), and then later inject the SR/CR immune cells at another site (e.g., intraperitoneally or into the abdomen).  This meant that the injected immune cells would have to migrate to the tumor and kill it at a distant site, all the while being in a normal mouse tissue environment.  Surprisingly, this strategy worked, and the established cancer in the normal mouse was killed by the SR/CR immune cells injected elsewhere.  Initially, the cancers on the back actually appeared to get slightly bigger after the immune cells were injected into the peritoneal cavity, but after a few days the cancer began to shrink.  The initial swelling may reflect an initial influx of active white blood cells into the tumor.  The cancers disappeared completely in two weeks.  As controls, other mice injected with similar immune cells from a normal non-resistant mouse showed no tumor shrinkage, and all of these control mice died at the expected time.  The surviving recipient mice were "cured" of their tumors (the tumor never recurred, even after a year – half a lifetime in mouse terms).  Thus, we can say with confidence that the killing of cancers by the SR/CR immune cells requires only the immune cells (and not something else) and is remarkably tumor-specific without causing significant harm to normal tissues.

Does A Specific Type of Cell Mediate Cancer Resistance?

The immune system in mammals is composed of many different specialized immune cells.  Some immune cells are specialized to recognize only foreign things and relay that message to other immune cells.  Some immune cells require prior exposure before they acquire the ability to protect the host.  Others are natural-born killers without needing any prior immunization.  Some immune cells are specialized to kill pathogenic cells by rupturing them, and others are specialized to eat enemy cells and kill them later.  If we knew which type of immune cell was responsible for cancer resistance, we might be able to design more efficient therapy for cancer patients.  One could enrich and expand this cancer-killing cell type to deliver maximum efficacy for cancer therapy.  One could also remove any potentially inhibitory cell types from the mixture, or immune cells that might potentially harm normal tissues, to achieve a better result.

This question can be addressed in two ways.  First, one or two types of immune cells can be removed from a mouse to see if the protection against cancer is still there in the resistant mice.  If removal of one cell type but not others could abolish the resistance, it could be guessed that this cell type was solely responsible for resistance.  When selective cell types were depleted from SR/CR resistant mice, however, no single cell type seemed to be necessary for cancer resistance.  When all immune cell types were depleted, however, resistance disappeared.

An alternative strategy is that specialized immune cells can be isolated to high purity and tested for cancer-killing activity.  After purification, one cell type, but not others, might be able to kill cancer cells.  We previously found that cells of the innate immune system, neutrophils, macrophages and NK natural killer cells, seemed to be the ones that attacked tumor cells.  If that were true, one might predict that one of these cell types might be the mediator of tumor cell killing, and that type of cell could be purified and tested alone.  Experiments showed that cancer killing could be also observed in the test tube (i.e., in vitro), rather than only in the intact mouse.  Both cells and tissue fluids were tested, and it was clear that it was the cells alone (not fluids) that were responsible for killing cancer.  However, when individual immune cells were isolated and tested for cell killing, we were surprised to find that no one cell type alone was needed, but that several types of innate immune cells from the SR/CR mice could kill cancer cells.  Similar cell types from normal mice tested in this same way were ineffective, showing that the effects of the mutation in the SR/CR mice was being expressed in all of these cell types, even in vitro.

Long-term protection in ordinary mice

The types of white blood cells we injected into the normal mice are thought to have a rather short lifespan (a few days or weeks).  Thus, we were surprised to see continued cancer resistance in the normal mice for months after they received SR/CR white blood cells.  There is more than one theoretical way in which this could happen, but one which we favor is that the purified white blood cells we injected contained a small fraction of stem cells, and that these gradually became part of the mouse's immune system.  Since we did these transfer experiments between SR/CR and normal mice of the same inbred strain, this is not entirely unexpected.  By performing these experiments using immune cells from a male SR/CR mouse and transferring them into a female recipient normal mouse, the injected immune cells could be identified later, because they contained a "y" chromosome.  In this way, we were able to show that some of the injected immune cells survived for a very long time and were probably involved directly in killing the distant cancer.

What Do These Results Mean For Human Cancer Therapy?

First, we should point out that in this mouse system the donor and recipient mice were both in the same in-bred laboratory strains of mice.  Thus, except for the SR/CR mutation, they are genetically identical.  Our transfer of immune cells between these mice is basically a transplantation experiment between identical twins.  If we tried this therapy in human patients, the transferred immune cells would probably not survive, since the donor and recipient would be very different genetically.

 However, these results show that the concept would work under the right circumstances.  For example, if we identified the gene, it might be possible to take immune cells from a patient and insert that mutant gene into those cells in the test tube, then give these cells back to the same patient; this would then perhaps allow the mutant immune mechanism to work to reject tumor cells without the loss of the immune cells due to transplant rejection.  However, this is a complex strategy that can have many potential problems.

A more important message from this work is that such a mechanism is actually possible in intact animals, and that a thorough understanding of the underlying molecular events could potentially lead to a new strategy for more specific cancer therapy.

There is second important message from this work. The fact that the cancer-resistant immune cells can specifically sort out cancer cells for rapid destruction suggests a fundamental difference between cancer cells and normal cells.  We can say with confidence that the killing of cancers by the SR/CR immune cells is remarkably tumor-specific without causing significant harm to normal tissues in an otherwise normal mouse.  For some reason, the immune cells from these special mice are capable of detecting these differences.  An important question is: “What are the common properties of different cancer cells that allow them to be distinguished by these special immune cells?”  While it is possible that cancer cells express something that activates these SR/CR white blood cells, it is also possible that these cancer cells may fail to inhibit these SR/CR white blood cells.  That is, the success of cancer growth may be through the ability of cancer cells to inhibit controls that normally limit the growth and spread of cells.  The SR/CR immune cells may ignore this common inhibitory function released by cancer cells, and treat them like any other out-of-control tissue.  We hope that by understanding this interaction between cells we can uncover clues to these underlying molecular mechanisms.

The Next Steps: Identification of the Mutation, Molecular Mechanism and Treatment of Endogenous or Naturally Occurring Cancer

Considerable work together with several collaborative groups has already been performed to identify the gene mutated in these resistant mice; however, this task is complex and still requires further work.  Other strategies to identify the altered molecular pathways include analysis of the genes expressed by purified populations of immune cells, comparing normal mouse cells to those from resistant mice before and after challenge with cancer.  Preliminary results of those experiments are very encouraging.  A further step is to document that this resistance mechanism works against endogenous cancer, as well as transplanted cancers.  Preliminary results from one mouse model of endogenous cancer are encouraging, suggesting that this mechanism is also active against spontaneous endogenous cancer.  Thus, through a variety of strategies, we hope to unravel the underlying molecular events responsible for this remarkable mouse, and eventually use this knowledge to design more effective therapies for human cancer patients.