Rumination 20. Clinical Trials and Tribulations

Rumination 20. Clinical Trials and Tribulations

The inanities and insanities of clinical trial protocols have caught up with me with a vengeance. I started the trial of the Cdk inhibitor SCH727965 in September at what at that time had been determined to be the maximum tolerable dose and I had a fairly severe immediate reaction to the drug - my blood pressure dropped and I went into the classic definition of shock which a rapid infusion of fluids quickly and easily corrected. Consequently the powers that be (i.e., the drug company) decided that the next time, three weeks later, I was to receive only half the dose. I also was given much earlier and more aggressive fluid support. At that dose and with the fluid support I experienced only the mildest of the expected side effects. Three weeks later, my regularly scheduled scans showed a noticeable drug effect on my tumors and the lower dose infusion the next day caused only the slightest sign of any side effects. The next treatment, three weeks later at the same lower dose had no side effects at all. My body clearly was getting used to the drug. I discussed increasing the dose with adequate fluid support and was told that once the dose had been reduced increasing it again was not permitted. The next set of scans (every six weeks) showed stable disease but no drug effect. Six weeks later (in January) my tumors had progressed and I was kicked off the study.

I have since learned that the established dose has been decreased further and indeed no side effects have been reported on this new dose. Of course, no success has been reported either. Administration of all drugs, and particularly new drugs, need to take into account something that all practicing physicians know (but too often ignore), namely that an individual's response to many drugs varies greatly. It makes little or no sense to try to determine maximum tolerated dose (MTD) on groups of patients where if even a small number of patients have serious side effects, everyone's dose is lowered. Dose escalation must be done on individual patients to be meaningful: Somewhere around 75% of the averaged MTD determined individually on the first group of patients is probably a good place to start dose escalation for the next group of patients. The low end MTD of the group is not necessarily, or even often, close to the optimal dose for the majority of patients. The Cdk inhibitor was showing every sign of working on me, but the drug company with the connivance of the FDA had my physician's hands tied and the lower dose, not surprisingly, had no effect. Is it any wonder that 80 - 90% of all clinical trials of new drugs fail? I have a suspicion that it is the lawyers (if anything goes awry we'll find someone whose fault it is, and, oh boy, will we make them pay) that bear the brunt of the blame, but the courts also bear responsibility for allowing the current state of affairs to develop and continue.

There is an interesting sidelight to all this. Drug companies love to complain about the cost of clinical trials and use this cost as an excuse to jack up the price of the few drugs that make it through the clinical trial process. Just consider how much money they would save if clinical trials focused on the clinical instead of the trial and considered therapeutic response from the very beginning. Furthermore, what seems never to be mentioned is that they bear only a small part of the cost. Much of the cost of the administration of the drug and the care of the patient during the clinical trial is passed on to Medicare and other insurance carriers, irrespective of the absurdity of the protocol, once the protocol has been rubber stamped by the FDA. "When will they ever learn, when ..."

None the less, unless I throw in the towel, I need to find a new study in which to participate. So, on February 11, I met with Dr. Shapiro in the morning and Dr. Hodi after lunch. Details of the considerations are discussed below the break. The upshot was unanimous agreement that an Hsp90 inhibitor was first choice of the way to go. (I have complained vociferously in the past about my interactions with Dr. Hodi. While I do not retroactively withdraw those complaints, I must report that what was true then is most certainly not true today. Our interaction was professional, most pleasant, highly instructive, and a much appreciated change from earlier days; I relish his highly knowledgeable input, positive tone, and offers of help.)

We all agreed that three classes of drugs were the most likely to be of benefit to me. Of these, the Hsp90 inhibitors were the most attractive for both theoretical and practical reasons. The theoretical reasons is that Hsp90 is a component of several of the relevant cellular signaling pathways. The practical reason is that no slots were available in the other trials for several months. Of the ongoing Hsp90 trials only two were practical for me in terms of travel and stay requirements. One is being run by NCI (National Cancer Institute, NIH) in Bethesda but would probably require relatively infrequent visits (it is an orally administered drug but I have not explored the details) and the other by Dr. Shapiro at DFCI that will require weekly visits for infusion. Because I have such high professional and personal regard for Dr. Shapiro and everyone on the staff of his unit, I asked to be considered for that Hsp90 trial, a Novartis drug, AUY922. I heard on Thursday (2/12) that Novartis has accepted me into the trial. I expect that the preliminaries, scans, EKG, etc., will happen next week. I note in passing how quickly things are changing in cancer therapeutics. Six months ago the consensus was that CTLA-4 inhibitors were the way to go and the Hsp90 inhibitors were speculative at best. I can't wait to find out what the drug of choice will be six months hence when Hsp90's stop working on me.

This is also the appropriate moment to talk about the genome sequencing on a sample of tissue removed from a metastatic lymph node in my neck, that was organized by my friends an N-of-One (http://www.n-of-one.com/), Drs. June Kinoshita and Jennifer Carter. The results are indicative of the current state of individualized genomic medicine. The analysis identified only mutations. Most of the mutations identified in my tumor code proteins with unknown function. Many cancer-related genomic changes are replications, not mutations, and the study that was done did not identify them. That is not to say that the study was not marginally useful. It allowed us to decide that Gleevec, a drug useful for c-KIT mutations but not replications would not help me. It also suggested that the available MET inhibitors, PF02341066, would probably not be useful in my case. (The results for the sample sent out for sequencing are not back.) So, while not as useful as such sequencing will be in the years to come, it was useful in reducing the field of choices for the next drug to try.

**************************************************************************************

I will go into some detail of my choice for the next trial, because it contains a cautionary tale of general applicability. It also demonstrates how very quickly the field of cancer drugs is moving.

Before my meetings on the 11th, I compiled a list of studies worthy of consideration. the list was

When I started my search, I thought that the CTLA-4 inhibitor with the unpronounceable name of Ipilumimab (Ipi) would be my first choice, because the two people I know who also have metastatic melanoma were on this drug and both had excellent long-term results with very few, if any, side effects. I did know that previous studies have demonstrated that no more than 8-12% of patients on Ipi received any benefit. What I did not know until I searched the recent literature [J. Weber, Overcoming Immunologic Tolerance to Melanoma: Targeting CTLA-4 with Ipilumimab (MDX-010). The oncologist 2008:13 (suppl. 4):16-25; and G. Q. Phan et al, CTLA-4 Blockade with Monoclonal Antibodies Metastatic Cancer: Surgical Issues. Annals of Surgical Oncology 15:3014-21 (2008)] was how frequent and serious long term side effects are with this drug. It turns out that benefit is usually attained only in patients who also have significant side effects. So, much to my surprise, that drug went from the top to near the bottom of my list. The cautionary moral of this story is how knowing a few patients can so mislead one from the realities of evidence based medicine.

Of course, what drug one chooses when there is neither a clear choice nor a clear promise of benefit, must perforce be based on a personal cost/benefit analysis. My cost/benefit analysis is based on my firm conviction that at my age quality of life is the overriding consideration; quantity counts for very little. Based on that premise, I reject any drugs with only short duration of benefit, if any, (which includes all approved drugs for melanoma) and that have the possibility of producing long-lasting (more than a few days) side effects. That eliminates both Ipi and Sutent because of their frequent serious, long-lasting, side effects. The Aurora A inhibitor is high on the list of possibilities because my prior good experience with an Aurora-B inhibitor (AZD-1152) that kept me stable for several months and had no appreciable side effects. The addition of a VEGFr inhibitor (to inhibit angiogenesis) to an Aurora-A inhibitor is a very interesting and desirable attribute of that study. Unfortunately, no slots for the study will be available for several months.

Melphalan infusion and recapture has the advantage of addressing my liver metastases directly, and those are the only metastases I have that are life threatening in the immediate future. However, this study, now phase III (NCT00324727) [J. F. Pingpank et al., Phase I Study of Hepatic Arterial Melphalan Infusion and Hepatic Venous Hemofiltration Using Percutaneously Placed Catheters in patients with Unresectable Hepatic Malignancies. J. Clinical oncol. 23: 3465-74 (2005); B van Etten et al., Isolated Hepatic Perfusion with Melphalan in Patients with Irresectable Ocular melanoma Metastases. Eur. J. Surg. Oncol. (2008) doi:10.1016/j.ejso.2008.07.004] involves monthly general anesthesia and multiple catheter placement, so I think it is best reserved for consideration at some future time when I become symptomatic, and hopefully, the results of the currently ongoing trial of this approach become available.

What makes an Hsp90 inhibitor so attractive is that Hsp90 is essential for the MET, KIT, and VEGF pathways, all of which are known to be involved in the proliferation of melanoma.

(http://www.novartisoncology.com/research-innovation/pipeline/AUY922.jsp?usertrack.filter_applied=true&NovaId=7852773814828258984 ). Currently there are several Hsp90 inhibitors in clinical trials. Of these, at least two are impractical for me because they are administered so frequently that I would have to move to Boston. Two studies of Hsp90 appear to be practical. The SNX-5422 at NCI in Bethesda (now there is a commute) which is an oral preparation and a study of AUY922 with Dr. Shapiro at DFCI that, after the first five weeks (one infusion per week), can be done as a day trip. (The first five weeks will require ancillary tests - lots of EKGs that will require overnight stays). So, AUY922 is going to be the next step. stay tuned.

(Previous chapters of my Ruminations can be found at http://upislandeggs.com/Ruminations.htm )