Using A Patient’s own Antibodies Against Fibrolamellar Hepatocellular Carcinoma:

A New Approach

Executive Summary: Scientists are increasingly convinced the immune system can help detect and eliminate tumors. The Simon laboratory, at the Rockefeller University, is seeking funds to help design a novel system of creating customized antibodies individualized directly from each patient to assist this process. Benefits may include: [1] finding FHC before it becomes inoperable; [2] vastly more specific delivery of chemotherapeutic agents that eliminates many of their harmful side effects; [3] improved real-time tissue evaluation during surgical procedures.

Let’s start at the beginning! How does the immune system find cancer?

The immune system is constantly surveying the cells of the body, looking for those that have changed either because of a viral or bacterial infection or a mutation such as found in cancer.

What happens next? How does the immune system attack a tumor?

When the immune system finds a changed cell, or clump of cells, it launches an attack by triggering those cells to undergo “programmed cell death,” (a process also known as apoptosis, if you prefer fancy terms). What happens during this process is the immune system gives the tumor cells a signal to commit suicide. Generally, all the cells follow these instructions. Occasionally, however, the tumor disobeys.

How can a tumor disobey a programmed cell death command?

The DNA of tumor cells is not stable: In other words, their DNA accumulates mutations at a rapid rate.  If, by chance, they have a mutation that disrupts the “cell death” pathway, then no matter how many times immune system tries to activate the suicide switch so that the tumor cell will kill itself, the cell death pathway does not respond. The immune system still recognizes the tumor cells but has lost the ability to kill them. In some cases, the tumors get even more sophisticated in their escape: They release signals that keep the “Killer Immune cells” away (yes, they are called killer cells). However, the immune system, and in particular the antibodies secreted by the immune system, still recognize the tumor.

What are we proposing to combat this?

We have good reason to believe that the immune system can recognize fibrolamellar hepatocellular carcinoma (as well as most other cancers). For instance, it is not uncommon for a patient’s lymph nodes (which are filled with immune cells) to swell to the size of golf balls. Usually, when the initial primary tumor is removed, much of the tissue is already dead and filled with immune cells that have been attacking the tumor. Unfortunately, in cases where fibrolamellar hepatocellular carcinoma has become a clinical condition, either the immune system is no longer killing the tumor cells or it is not killing them fast enough.

The approach of the Simon lab at the Rockefeller University is to take advantage of the fact the immune system can still recognize these tumor cells, whether or not it is effective at triggering a successful suicide. We have developed a technology in which, using a few drops of blood, we can purify a patient’s (or animal’s) antibodies, then attach or “tag” them with a harmless, natural substance that glows. The glowing antibodies can then be seen under the microscope. They can also be injected back into the patient’s body, where they can be seen with a scanning machine such as a CT or MRI. Because these antibodies bind to the tumor cells, they can be used to search for the existence of fibrolamellar hepatocellular carcinoma in the living human body. There is currently no similar diagnostic tool in existence.

We have devised a method for tagging antibodies not just with a harmless substance but also with a toxin. This opens the door to a host of therapeutic purposes. In short, every patient would have a set of personalized tagged antibodies, created by his/her own immune system, with an assist from our lab.

What are the diagnostic uses for a patient’s own customized antibodies?

1) During resection of fibrolamellar hepatocellular carcinoma, the surgeon is in constant contact with the pathologist. Each piece of tissue that is removed is examined by the pathologist to determine if there are sufficient margins around the tumor that are clean. The use of antibodies could quickly help ensure – during the course of surgery - that the margins were clear.

2) Current scanning techniques may report on spots that are “hot” in various locations of the body. However, a positive signal does not always equal a metastasis of the tumor. Labeled antibodies would be a far more specific diagnostic tool..

3) Labeled antibodies will allow physicians to probe for sites of fibrolamellar hepatocellular carcinoma that may be too small to be detected by other means. The hope is that when tumors are smallest, they are easiest to fully remove with minimal damage to the body.

What are the therapeutic uses for a patient’s own customized antibodies?

If the technique of using an antibody to deliver a glowing substance to the tumor works, then we can use the antibody to deliver other things to the tumor such as a poison. Just because some tumors evade the suicide command does not mean that we cannot still use the immune system to target a toxin to the tumor. The ability to target toxins means that chemotherapeutics could be used at significantly lower concentrations and thus reduce, or perhaps eliminate, the undesired side-effects.

Why is there such a need?

Unfortunately, fibrolamellar heptocellular carcinoma often portends a poor prognosis. One factor that may contribute is that it is often detected late in its course. A second factor is the absence of good markers for fibrolamellar hepatocellular carcinoma. The ability to use a patient’s own blood as a marker for the tumor would add strength to an evaluation of the margins of a resected tissue. Further, it would allow an evaluation of whether there are other foci of the tumor elsewhere in the liver that may not be obvious yet with other scans. It would additionally allow detecting potential metastases elsewhere in the body – hopefully while they are still easily resected.

Has this approach been tried before?

Yes, but earlier efforts were not successful. Our approach is different in several critical ways:

To explain how, however, it is necessary to give a bit more detail about how the immune system works. When there is a change in cells of our body as a result of an infection or a malignant transformation, the cell tries to signal this information to the immune system. Specifically, the cell chops up one or more of the offending proteins, and presents fragments of the protein on the surface of the cell as a warning flag. Why? If this cell is detected and destroyed, then the rest of the organism can be protected. In the case of tumors, these fragments are called tumor associated antigens. Using antibodies from patients, scientists have been able to characterize thousands of tumor associated antigens. It is not clear what changes during the development of tumors result in particular proteins being expressed on the surface of cells. However, there is growing evidence that a number of tumor associated antigens can be diagnostic for certain forms of cancer.

So far, over a thousand human tumor associated antigens have been identified. This has led to significant efforts to identify and characterize the tumor associated antigens for the purpose of both detection and therapeutic treatment. Some groups have tried to generate specific antibodies to these tumor associated antigens. Others have taken tumors out of patients, grown them in a dish, and then use them to make antibodies in “humanized” mice for similar purposes of detecting or treating the tumor. Unfortunately, neither approach was particularly successful and they were dropped. However, there were at least two major reasons why we believe they may have failed:

First, once you take a tumor out of a patient and start growing it in a dish, its behavior changes dramatically... as does the molecules that it is making. Thus, it will elicit different antibodies than the same tumor inside of the patient. Second, the cells of each person present different fragments of proteins on their surface to stimulate antibodies. Two different people may have the same kind of cancer, but the cells of their body will present to the immune system slight differences in their profile. Thus each person’s immune response is different. There is nothing as potentially powerful as a patient’s own immune response to a tumor. That means that each response has to be personalized to that patient: you take a sample of the patient’s own blood, link it to a reporter or a poison, and use it back in that individual. Such an approach is much more likely to succeed.

What is the evidence that favors our hypothesis? There have been no studies detailing the antibodies generating in patients with fibrolamellar hepatocellular carcinoma. However, information on work on hepatocellular carcinoma (HCC) is instructive. Autoantibodies were detected in the blood of HCC patients prior to the development of tumor, but is not clear at what point during tumor development these antibodies are produced and what their course is over the various stages of tumor development. In HCC, the transformation of liver cells from noncancerous to cancerous is associated with a significant antibody response.  Studies have identified numerous tumor-associated antigens that are associated with HCC. However, each tumor associated antigen is not found in every HCC tumor (the presence of each tumor associated antigen found in each tumor has been shown to range from 9 -22%), and therefore is not a reliable antigen in individual patients. As with other cancers, it is unlikely that all of these antigens have been identified. While antibodies to individual tumor associated antigens have been found to be weakly diagnostic of a tumor, panels of numerous tumor associated antigens are more successful.

In this project, rather than purify and characterize individual antibodies to specific tumor associated antigens, we will use the spectrum of antibodies generated by an individual animal to detect the tumor in that same animal. It is our working hypothesis that some of the variability in the success of using antibodies to tumor associated antigens is the result of both variations in the antigens amongst same-type tumors, as well as variations in individuals’ immunological responses. We will label all of the antibodies in the serum of an animal to test if they can distinguish tumor from non-tumor tissue from that same animal.

Regards,
Sandy Simon
Professor, Head of the Laboratory of Cellular Biophysics
The Rockefeller University
1230 York Avenue
New York, NY 10065
212-327-8130 (voice)
simon@rockefeller.edu

Dr. Simon did his undergraduate work at Princeton University in neuroscience and received his Ph. D. from NYU School of Medicine in Physiology and Biophysics.   Dr. Simon came to The Rockefeller University to do post-doctoral studies with Gunter Blobel. The results of some of their work was highlighted in Dr. Blobel’s 1999 lecture for his Nobel Prize and led to the appointment of Dr. Simon to the faculty of the Rockefeller University. Dr. Simon’s work has ranged from studying single molecules moving in cells, single machines inside of the cell as they operate, and studying single tumor cells as they metastasize in the mouse. His most recent publication, which appeared in the journal Cell over the summer, studied how cells secrete. Last year (2008) one of his articles in the journal Nature, which presented the first images of a virus (in this case HIV) assembling in a living cell was selected by Discover Magazine as one of the top 50 science stories of the year.

Dr. Simon's research is indeed exciting on many fronts but he needs our ongoing financial support to continue this project...Your contribution to  "DeLiver a Cure" will help keep this project moving forward.