SLIDES & TRANSCRIPTS
Wednesday , June 20

CHEMOTHERAPY/ALTERNATIVE DELIVERY METHODS SECTION - GROWTH FACTORS AND SIGNAL TRANSDUCTION INHIBITORS

Paul Bunn, Jr., MD

Small lung cancers offer the opportunity to study new agents for their ability to reduce the size of the primary tumor to prevent recurrence and to prevent second primary tumors or SPTs. What is the rationale for this? Growth factor receptors are overexpressed in preneoplasia and receptor antagonists inhibit lung tumorigenesis in animal models and reduce the size of human lung cancers in animal models. In humans growth factor inhibitors are given orally, are safe and produce responses in advanced disease.

TOP

So, the idea came from Sporn and Todaro several decades ago that even during the process of pathogenesis of lung cancer and in their progression we have local growth factors that cause the metaplasia, hyperplasia and dysplasia and then develop an autocrine system where the tumor is both producing a growth factor as well as expressing the receptor.
In terms of lung cancer growth factors most of them are peptides. Primarily it is the epidermal growth factor family for the non-small cell lung cancers and the neuropeptides such as bradykinin and GRP for small cell tumors, although certainly a significant fraction of non-small cell tumors express peptide receptors.

TOP

So, what does this look like during the pathogenesis process? This slide is via Wilbur Franklin who is in the room here. The normal epithelium is 1 or 2 cells thick as you know and in every epithelium it is the basal layer where cells divide and replenish themselves and so expression of the EGF receptor in the normal epithelium is confined to the basal layer, but early in the pathogenesis of lung cancer and exposure to carcinogens you can see that the epithelium becomes thicker and after you get to hyperplasia, metaplasia and dysplasia not only is the lining several cells thick, but the nuclei are dysplastic. This shows the EGF receptor staining in a dysplastic lesion, you will be hearing about another lesion called angiogenic squamous dysplasia where there is actually angiogenesis going on in the dysplastic epithelium which is also a common precursor and is the subject of the next talk. But once again you can see that the EGF receptor is playing a prominent role in not only the growth of the dysplastic cells but in the angiogenic area, and obviously advanced tumors widely express the EGF receptor. So, that is sort of the target validation.

TOP

Now, an interesting thing about all these growth factors is we originally thought that what they did caused proliferation. Now, in general for the EGF family ligand binding to the receptor causes dimerization of the receptor, and it is the dimerization that leads to phosphorylation of the internal portion of the receptor. Originally most people concentrated on the ras, raf, map kinase pathway which was believed to lead to proliferation. Oh, what a great slide here. Maybe you can see that. It is on my projector, anyway. That is sort of the ying, the proliferation part, but there is a yang to all these pathways which is the role in apoptosis, and actually when the receptor is phosphorylated you get AKT through PI3 kinase, and that actually inhibits apoptosis. Now, antagonists have the opportunity to both prevent this pathway which will lead to increase in apoptosis and inhibit this pathway which leads to decreased proliferation.

TOP

Slide 6: EFG Receptor Blockade Targeted Therapies

So, we have two ways primarily of interfering with this pathway that have been developed clinically. That includes the monoclonal antibodies that bind to the receptors originally mouse antibodies, then chimeric antibodies and now actually human antibodies made in a mouse with human immunoglobulin genes, and then we have the small molecule receptor tyrosine kinase inhibitors. Some of these are very specific for EGFR. Some bind to other receptor tyrosine kinases. Some have reversible binding to the receptor and some have irreversible binding to the receptors. I am not going to talk about immunotoxins. They have, also, been developed but only preclinically so far and I am not going to talk about combinations at the moment.

TOP

What happens with these agents in preclinical models? We have a target. We understand the signal. What happens in the preclinical models? This is ZD1839, one of the small molecule inhibitors. This is the antibody against lung cancer cell lines in vitro, and you can see that the small molecule inhibitors in vitro are better at inhibiting lung cancer growth but there is a fairly wide sensitivity to various lung cancers to these agents. I am going to point out two cell lines here. One is a cell line called H226 which is inhibited by both agents although it is inhibited more by ZD1839. That particular cell line expresses both EGFR and interestingly enough also expresses HER-2, and then we have A549 which is also inhibited by both agents, also, inhibited more by ZD1839 and this particular cell line has about the same amount of EGFR but does not have HER-2.

TOP

So, if we look at the signalling effects of these compounds in vitro this is the H226 that expresses both receptors. This is A549 that is expressing only EGFR but not HER-2 and this is treatment with C225, and this is treatment with IRESSA and let us just look at the 4-hour time point. First of all you can see that both cell lines express EGFR and you can see that phosphorylation of the EGF receptor is blocked in both cell lines by both ZD1839 and by C225. Here is again the 4-hour time point with both agents; phosphorylation of the EGF receptor is blocked.
In terms of the ying part of the pathway the map kinase part of the pathway both agents are inhibiting map kinase. Interestingly, although ZD1839 is more effective and although there is a greater effect on 226 than there is on A549, the phospho map kinase is inhibited more in A549, and it is inhibited perhaps more by C225. So, that would go against the greater growth inhibition in this cell line and the greater growth inhibition caused by this. If you look at phospho stat 3 or phospho AKT there is a different result and that is that there is greater inhibition from both agents in 226 consistent with the greater effect in 226, and there is a greater effect of ZD1839 than there is of C225. So, it may be actually that stimulation of apoptosis is more involved in the actual growth inhibition than the inhibition of proliferation

TOP

With respect to preclinical animal models both of these agents inhibit the growth of human lung cancers in athymic mice. This is from Memorial Sloan- Kettering, a study of Dr. Soradnick, and when combined with standard chemotherapy greater growth inhibition is achieved, and in fact with a number of these agents you can completely prevent any tumor growth whatsoever for as long as you deliver the agent. In combination it is possible although not totally proven that you might actually cure these animals of their tumor.

TOP

Now, what about people? A surprising finding was that the earliest studies of these agents in people with highly refractory disease actually had objective responses and this really is a response rate of 2 of 16 or just 12.5% although there were two minor regressions. In the study of OSI-774 another small molecule inhibitor the response rate in pretreated patients was 11%. So, consistently there is between 10 and 15% response rate in heavily pretreated patients which was actually higher than any cytotoxic chemotherapeutic agent, and these things produce even higher responses in less heavily pretreated patients.
Randomized clinical trials, several with over 1000 patients, have been completed. So, there are several thousands of patients who have received these agents. So, in terms of safety there is a tremendous database for these agents.

TOP

The major problem in terms of toxicity is a maculopapular acneiform rash which occurs primarily on the face and upper body, less so on the extremities although it can occur on the extremities. Really a major issue with these agents is whether the patients will be compliant. Obviously this rash is not life threatening. It goes away when you stop the drug, but certainly for chemoprevention a major issue is will this acne prevent compliance with these agents as a therapy for either chemoprevention or early lung cancers.

TOP

This is just to say that these 1000-patient trials that gave conventional chemotherapy with or without these agents are actually closed now and this one is about to actually start.

TOP

So, a chemoprevention study done by an NCI supported group called the Lung Cancer Biomarkers and Chemoprevention Consortium which includes the SPORE institutions plus Duke, Vancouver, Moffitt and a couple of other institutions, will be starting any day. In these high-risk patients, namely patients who have had a prior cancer, 30 pack years of smoking and sputum atypia will be stratified by the site of their original tumor and their smoking status, current versus ex-smoker, and will be randomized to placebo or to the EGFR antagonist, ZD1839 in a two-to-one randomization. The intervention will last 6 months. There will be bronchoscopy before and after the intervention, and there is a crossover for progressors who receive placebo.

TOP

Another target is downstream of the receptor. Certainly two common pathways involved actually in both types of peptide receptor pathways include ras and PKC. I am not going to talk too much about PKC inhibitors at the moment, but will talk about ras.

TOP

Originally to become activated, ras from its intracellular cytoplasmic location has to bind to the cell membrane. To do that it requires a modification called farnesylation and so farnesylation inhibitors to prevent activation of ras were developed.

TOP

Of course, we know that these are quite effective as farnesylation inhibitors. Major questions have arisen as to whether inhibiting farnesylation of ras is actually the key protein involved. Some of that is shown here. This slide came from Alex Adjei and so I may screw this up and Alex will correct me if I get this wrong. Anyway a number of farnesylation inhibitors have been developed, and there are actually some targets for in vivo studies to determine the optimal dose and concentration of these compounds. This is some in vitro work. This is a non-small cell lung cancer cell line treated in the uncontrolled with an inactive agent or with two farnesyltransferase inhibitors, and you can see here that HDJ2 is farnesylated. The farnesylation is inhibited by both of the inhibitors in all of the cell lines, and prelamin-A - actually Alex can tell you how this works - is actually working as well. It is just the opposite here where we get farnesylation and you get increases, but ras and P21 don't seem to be much affected and these compounds do inhibit the growth of the lung cancer cell lines.

TOP

This can be done not only in Western blot analyses but you can actually develop immunohistochemical assays. These immunohistochemical assays will work in cell lines and it doesn't matter whether the cell line is growing rapidly or even in the confluent stage where it isn't growing rapidly you can get an effect with these compounds and you can measure the effect of the compounds in an immunohistochemical assay.
So, in a person you can do a skin biopsy or a buccal scraping and see that you are having the effect that you want.

TOP

The question is do these agents then do anything? They have been given to many hundreds of patients now. This slide was given to me by Fadlo Khuri. This was a study in lung cancer patients in which patients were given the Schering compound plus Taxol because again of in vitro data suggesting synergistic or additive effects, and of the patients with non-small cell lung cancer there were 12, and this is a combination study, Taxol and the FTI inhibitor.

TOP

In the first 12 patients there were six partial responses and three minor responses.

TOP

You can see that there is a dose-response effect but looking at prelamin or DNAJ you can as I mentioned, determine what is the optimal dose to affect the target, and you can see there is a dose-response effect and you are getting a considerable effect at doses which are tolerated in people.

TOP

And in this early study, again, this is head and neck not lung but a similar pathogenesis, responses have been observed even in the Phase I trials.

TOP

Again there are a number of these agents. Some of these are oral; some of these are IV, basically most of them cause diarrhea as their major side effect. Some have skin toxicity. Some cause myelosuppression which tends to be mild, but basically at doses that hit the target these agents are relatively well tolerated.

TOP

Once again this led the Lung Cancer Biomarker and Chemoprevention Consortium to develop chemoprevention trials, the same group of patients, prior cancer, 30 pack years, smoking atypia, same study design, that is randomized to placebo for 6 months or the active compound in a 2 to 1 randomization for 6 months, rebronch, look at the biomarkers. The main marker by the way for these studies is histology and the secondary biomarkers, the main one is KI67, but there is a whole panel of biomarkers. So, those compounds are in chemoprevention trials and potentially could be used in this population.

TOP

Now, I don't have time, but we could have actually had a talk on each of these other classes of agents. The GGTIs may be as important as the farnesyltransferase inhibitors. Combinations of these have been tested but seem to be quite toxic in combination. There is a PKC inhibitor. It is actually an antisense compound which is in Phase III clinical trial, perhaps not as sexy for this application because this is a continuous intravenous infusion. M-TOR inhibitors are, also, signal inhibitors. This is the mammalian target of rapamycin and the company that makes the lead compound found several advanced refractory lung cancer patients responded in Phase I trials although they are not developing the compound in lung cancer.
Several raf inhibitors, cell cycle inhibitors proteosome inhibitors and neuropeptide inhibitors are in preclinical models and are potentially useful for this group of patients.

TOP

Then the question becomes, and this is my last slide, is if we have all these combinations and preclinically they work better together in combinations when should we do combination studies in people? I would submit that probably for each agent we need target validation. We need preclinical and clinical efficacy studies. We need safety, and we need convenient schedules.
Certainly it would be best if we are going to do a combination in people if there is in vitro synergy, and it would be best if there are preclinical animal models showing additive or synergistic effects without an increase in toxicity. Certainly one would want to have Phase I combination safety data and some evidence that you are getting a really good effect. I am not sure that any of the combinations are ready for prime time but certainly in the future I think that some of these will be given in combination not just singly.
So, I would be happy to answer any questions.

TOP