The Smart-Pill Solution

[gdpawel]

Monoclonal antibodies like Herceptin, are large molecules that attach to specific proteins on the outside of cancer cells and do not have a convenient way of getting access to a large majority of the targeted cells on the inside, which are protected from the drug. Plus, there is multicellular resistance, the drugs affecting only the cells on the outside may not kill these cells if they are in contact with cells on the inside. The cells may pass small molecules back and forth. This would be a very good reason Tykerb, a small molecule drug, may be much better than Herceptin.

However, each of these new targeted drugs are not for everybody (just as conventional cancer drugs are not for everybody). Even when the disease is the same type, different patients' tumors respond differently to the same agents. As the saying goes, "don't throw out the baby with the bath water." If a drug works extremely well for a certain percentage of cancer patients, identify which ones. If one drug or another is working for "some" people (not average populations), then obviously there are others out there who would also benefit.

The study of cell function analysis tells us that even when the disease is the same type, different patients' tumor respond differently to the same agents. Herceptin (or any other large molecule targeted drug) may be more beneficial to some patients than Tykerb, Sutent, or any other small molecule targeted drug.

Whatever the percentage of patients benefit from these drugs, the point is, "targeted drugs are not for everybody." Pre-tests can help identify the individual cancer patient the drug works extremely well for, or it can tell that the drug is resistant. This could be Tykerb, Tarceva, Iressa, Sutent or Nexavar, because of being small molecule drugs. It is important to "personalize" cancer treatment, and this can be accomplished by "testing the tumor first."

There are huge economic problems here. Pharma cannot make drugs unless they can realize a profit. The ordinary trial system will not suffice if we are to encourage new drugs for restricted numbers of patients. More and more physicians and patients are turning to individualized therapies to treat cancers. Without individualized testing the efficacy of these drugs, it's difficult to determine which drugs are best for patients who don't respond to standard therapies.

[Sandy in Silicon Valley]

Dear gdpawel,

I do understand the importance of pre-testing for the likelihood of targeted therapies, like Herceptin and Tykerb, to be effective against one's bc cancer cells.

However, are the extant pathology dx tests accurate? When/if a cancerous mass contains more than one type of bc cell (my understanding is that this is, or will be, the case in most bc detectable masses before/during/after treatment), how do such tests determine which of the bc cells will die, and which will survive and go on to reproduce, in response to a particular targeted treatment?

For example, I have a "spot" in the right caudate of my brain, adjacent to the basal ganglia. This "spot" (then .5cm) was irradiated with CyberKnife in July 2005. It was undetectable in the following 3 brain MRIs. However, in Oct. 06, the "spot" showed up again, and has seemingly increased in size approximately .2cm unidirectionally, with each subsequent brain scan since then.

It could be progressive radiation necrosis (the neuro- oncologist I consulted believes so), it could be regrowing cancer cells (each possibility due to opposite scenarios - too much SRS radiation in 2005, or too little). It is too deep to biopsy.

I am on Tykerb and Herceptin currently - w/o concommitent chemo. According to my oncologist, Tykerb has only shown about 28% effectiveness against extant brain mets or growth of new lesions, in HER2-neu (with hx of brain mets) patients taking it.

Due to unpleasant side effects, I would like to stop taking Tykerb, if I knew it wasn't effective vs. my brain mets, since the Herceptin appears to be still effective other than past the BBB.

How would my bcmets be tested to predict if Tykerb was or wasn't likely to be effective in lengthening the time-to-recurrence and/or survival for me?

It's my understanding that more and more longterm radiation necrosis, some of it serious-to-fatal, is showing up in bcmets patients with hx of irradiated (WBR or SRS) brain mets. How would the targeted therapies affect such side effects, if at all?

Your knowledge is extensive, and it seems like you might have been a biochemist, working with (or studying about) cancer cells in vitro. From what I've read and heard, in vitro, and even in vivo, studies only very rarely (<10%) turn out to have relevance for human subjects, due to a number of factors.

So, what IS a "smart" pill, and how would one know that it is "smart" about one's own bc pathology?

Thanks,
Sandy in Silicon Valley

[gdpawel]

I've been a cancer patient advocate and a student of cell function analysis for a number of years, like anyone would have an interest in molecular science or biological science. I do not have any financial conflicts of interest raising the awareness of this technology. My point with respect to this systematic procedure is to educate patients and others that such techniques exist, and might be very valuable. I get nothing out of my endeavors except the satisfaction of knowing that I've helped to increase the knowledge of informed consent. I get no pay, no lectureships, no junkets, not even any free meals.

Lee Newcomer, former chief medical officer and currently an executive with United Health Group, stated at the 12th annual conference of the National Comprehensive Cancer Network (in regards to the new "targeted" drugs), that "Avastin improves outcomes in about 20% of patients, but we have no idea which cancer patients will benefit from a course of treatment. Because Avastin is included with numerous drug cocktails, it costs $354,000 per year of life extended with Avastin because of today's 'cookie-cutter' approach to chemotherapy. You don't know in advance who is going to respond."

Although these "targeted" therapies are initially effective in certain subsets of patients, the drugs eventually stop working, and the tumors begin to grow again. This is called acquired or secondary resistance. This is different from primary resistance, which means that the drugs never work at all. Initially, tumors have the kinds of mutations in a certain gene that were previously associated with responsiveness to these drugs. But, sometimes tumors grow despite continued therapy because an additional mutation in the gene, strongly implies that the second mutation was the cause of drug resistance. Biochemical studies have shown that this second mutation, which was the same as before, could confer resistance to the mutant gene normally sensitive to these drugs.

In pre-testing with "cell-death" assays, there is no growing of cells as in the old "cell-growth" assays that you may only be aware of (which unfortunately, a number of old school oncologists are only aware of). Cells are taken fresh "live" in their three dimensional, floating clusters (in their natural state). Cells are cultured in conical polypropylene microwells for 96 hours to increase the proportion of tumor cells, relative to normal cells.

A Functional Tumor Cell Profiling in cell culture assays is the one test in oncology that can have the greatest immediate potential to improve therapy selection for "individual" cancer patients, not the "average population." Data from studies demsonstrate close correlation between prospective predictions of drug activity and patient chemotherapy response and overall survival. When individual patients are treated with drugs "active" in the assay, they have vastly superior response and survial rates than when they are treated with drugs which are "not active" in the assay.

The new paradigm of requiring a companion diagnostic as a condition for approval of new targeted therapies still does not guarantee that a drug will be effective for an individual patient. Companion diagnostics that were approved often have been mostly or totally ineffective at identifying clinical responders (durable and otherwise) to the various therapies. Nor can they, for any patient or even large group of patients (like in a clinical trial), discriminate the potential for clinical activity among different agents of the same class.

When the decision is made to treat a patient with chemotherapy, most patients are treated with a combination of drugs. No available gene-based test can discriminate differing levels of anti-tumor/anti-microvascular activity occurring among different therapy drugs. The "functional profiling" method of a cell culture assay differs from existing DNA and RNA tests in that it assesses the activity of a drug upon combined effect of all cellular processes, using several metabolic and morphologic endpoints. Other tests, such as those which identify DNA or RNA sequences or gene expression of individual proteins often examine only one component of a much larger, interactive process.

Conventional chemo treatments try to kill all cancerous cells (along with non-cancerous cells to boot). The whole forest of cells. The new targeted drugs go after a "pathway" within or on cancerous cells. Hence the "trees" instead of the "forest." With "cell-death" assays, the "forest" is looked at and not just the "trees." There are many pathways to altered cellular (forest) function (hence all the different "trees" which correlate in different situations). The "functional profiling" technique measures what happens at the end (the effects on the forest), rather than the status of the individual trees. Cancer is a complex disease and needs to be attacked on many fronts.

Cancer therapy needs to be thought of "outside the box" with "personalized" treatments for "individual" patients, and requires a combination of novel diagnostics and therapeutics. If "some" drugs are working for "some" people (not average populations), then obviously there are others out there who would also benefit. Who are those that would benefit? All the more reason to pre-test the tumor first. The "smart" pill is the one (or combination) that goes after all the pathways.

Literature Citation:

Eur J Clin Invest 37 (suppl. 1):60, 2007
PLoS Medicine, February 22, 2005