BAY 43-9006 (Sorafenib)

BAY 43-9006 is a targeted drug in clinical development and is only available in clinical trials for people who’ve already tried the standard treatments. There is little published data, yet the results reported so far while preliminary and only presented at meetings, are definitely promising, and unusually so. In addition to published data, a quite a few patients on theKIDNEY-ONC E-Mail List have been in the phase II trial, and a surprising number are reporting their tumors are shrinking and in fact we realized something was up well before any data was publically presented

BAY 32-9006 was first developed by Onyx Pharmaceuticals. Onyx subsequently partnered with a large and well known drug company, Bayer (Bayer is the “BAY” in BAY 43-9006) to complete development of the drug.

BAY 43-9006 is an oral drug with moderate to low side effects for most patients. As cancer therapies go it’s apparently pretty easy to take for most. A few patients do have serious side effects although these appear to be treatable and not actually life threatening.

How It Works

BAY 43-9006 is a “targeted drug” specifically engineered to inhibit something called the RAF kinase within the cancer cells. RAF in turn is part of the RAS oncogene pathway. RAS is a gene which drives cell division and is overexpressed in many cancers, including RCC.

It turns out that in addition to targeting RAF Kinase, BAY 43-9006 also inhibits the VEGF and PDGF receptors on blood vessel cells. I believe that these effects were accidental rather than designed. Both of these receptors are also kinases so that it’s not too surprising a drug which inhibits one kinase might also inhibit others. Clear cell kidney cancer is well known for vastly over-producing VEGF, which in turn stimulates blood vessel growth which supplies the tumor with oxygen and nutrients. Most clear cell RCC has a defect in what is known as the VHL (Von Hippel-Lindau) gene which causes the cell to think it is short on oxygen and to pump out huge amounts of VEGF. BAY 43-9006 interrupts this signal on the blood vessel cells. I believe that this inhibition of VEGF is probably the main reason BAY 43-9006 has had much greater effect in kidney cancer than in other cancers.

Onyx has a diagram on its BAY 43-9006 that explains how BAY 43-9006 works better than I could. Also, CancerQuest, an amazing site which explains the biology of cancer to laymen, has a page on oncogenes and if you look there you’ll see a link to a page just about the RAS oncogene.

The Data

The data in kidney cancer is from a phase II trial with an unusual design called “randomized discontinuation.” The randomized discontinuation design is intended to make it possible to tell if cases of “stable disease” is actually being caused by treatment or whether they simply reflect natural variability in tumor growth rates. In this design, patients who have stable disease after the first evaluation are randomized to recieve either placebo or to continue on treatment. In this case the assignment is double blind – neither the patient nor his doctor knows whether he is on drug or treatment. Once progression is seen, patients on placebo are permitted to cross-over and resume treatment with the study drug.

This particular phase II study was orginally open to patients with any kind of cancer resistent to treatment, but particularly to patients with colon cancer. But after it was observed that colon cancer patients weren’t benefiting and that a surprising percentage of kidney cancer patients were, they investigators changed direction and changed the trial to accept kidney cancer patients only. As we observed many people responding on the KIDNEY-ONC E-Mail List many more of us signed up for the trial. The list actually helped power rapid accural to this study.

The most recent data is from the 2004 ASCO meeting [Ratain 2004] and this is the most important source of data we have, but I am also relying on a presentation from October 2003 at the 2nd International Symposium on Signal Transduction Modulators in Cancer Therapy in Amsterdam. My information on that is based on personal communication with Dr. Ratain but the interpretations are my own.

Key Results:

  • 89 Evaluable Kidney Cancer Patients
  • 15% Partial Responses (More than 50% shrinkage) but no complete responses
  • 27% Minor Responses (25 to 49% shrinkage)
  • 50% Stable Disease at first evaluation (12 weeks)
  • 8% Progressive Disease at first evaluation (12 weeks)

Time To Progression: It is difficult to estimate the time to progression for the treated patients because half the patients in the randomized part of the trial were taken off therapy. At this point these patients haven’t been unblinded for data analysis so we can’t just remove them from the analysis. Interestingly though, the curve for progression in the group that was randomized shows a rapid progression of about half of these patients combined with much slower progression for the other half. This suggests, as Dr. Ratain implied during his presentation, but does not prove that the fast progressors are mainly the patients randomized to placebo while the slower progressors are those randomized to continue treatment. It looks to me as though the effect could be quite striking and I would guess that when the discontinuation data is properly analyzed there will be a large and highly statistically significant effect. If you look at the progression for the non randomized patients (also shown in the presentation) and combine it with the randomized curve assuming the rapid progressors there were on placebo it looks like the median time to progression might be close to a year (very roughly). Since other studies have shown a time to progression in untreated RCC of about 2 to 3 months, this would be a large and important improvement.

Durability of Responses: I was disappointed that the 2004 presentation didn’t specifically include data on durability of response though the time to progression data surely suggests some responses were durable. From the October 2003 presentation the time on therapy for partial responders who’d reached an evaluation at 18 weeks (IN WEEKS) was:

36+, 36+, 36, 24+, 24+, 24+, 24+, 24, 18+, 18+

A “+” sign means the responder was still in response at the time listed. No plus sign means the patient had significant tumor growth and was removed from the therapy at the given time. This means that 8 of these 10 responders were still in response at the last measurement.

Side Effects: 5% of patients had their dose reduced due to side effects, and 18% had their treatment temporarily stopped due to side effects. No patient had to actually stop treatment. The most common side effects were rash, hand and foot syndrome (which means pain, swelling, redness of the hands and/or feet), fatigue and diarrhea. Most were mild to moderate (grade 1 or 2) but some were more serious (grade 3). None were life threatening. The doctors felt that overall this drug could be used for long term administration. The October 2003 presentation gave a little additional information. According to that presentation 36% of patients had no significant side effects at all, while somewhere between 35 and 40% had at least a grade 3 side effect which means pretty significant. Again, even so called mild side effects (grade 1 or 2) such as nausea or diarrhea can be wearing if they occur over an extended period of time. Based on reports from the KIDNEY-ONC E-Mail List I suspect most of these side effects were not that relentless.

Special Situation and Warning

Unfortunately, this treatment is currently only available in a phase III randomized trial in which half the patients are intentionally given only a placebo (inactive "sugar pill"). The study is double blind, which means neither the doctors nor their patients know who is getting the drug and who is getting a placebo. Moreover, patients on placebo are not allowed to switch to receive the drug - not even after their disease worsens. Instead, they will be told whether they were on drug or placebo and can seek other treatment or clinical trials. For patients, this trial is close to worst case scenario. Specifically this trial design reduces the promise by exactly 50% compared to one in which you would actually be guaranteed treatment.

Although you may be able to get real treatment after your tumors grow while being treated with placebo, in addition to not actually getting the drug, the delay in getting actual treatment reduces your chance of being well enough to qualify for another trial or treatment. The total delay is more than you might think. First, the trial requires that your tumors grew with a standard treatment, either IL-2 or Interferon based before you can enroll. Then you have to wait 30 days before you're eligible to start this treatment. Then you have to wait till your tumors grow significantly again before you find out you got the placebo. Then it takes at least some time to decide on and start another treatment. Added together, that's a lot of your precious time spent on a sugar pill.

I believe randomizing patients between a placebo and an obviously promising drug in a disease which is almost always fatal without effective treatment is both wildly unethical, totally unnecessary, and just plain cruel. I am working with a patient group to get this trial changed, but so far not one iota of progress has been made, and unfortunately, I cannot offer much hope that changes will be forthcoming.

I predict there will come a time when intentionally giving placebo to patients with refractory metastatic cancer will be understood to be an ethical stain which, like the stain of notorious abuses of the past associated with medical research, will be impossible to erase. I predict the time will come when all will be astonished and horrified that anyone could have conducted experiments on patients for whom the stakes are literally life and death in which half were quite intentionally offered no help and no hope.


Ratain MJ, Flaherty KT, Stadler WM, O’Dwyer P, Kaye S, Xiong H, Patnaik A, Gore M, Lee RJ, Eisen T
Preliminary antitumor activity of BAY 43-9006 in metastatic renal cell carcinoma and other advanced refractory solid tumors in a phase II randomized discontinuation trial (RDT)
Proc Am Soc Clin Oncol (ASCO Meeting Abstract) 2004 23 : Abstract 4501 [You can find this abstract on the ASCO website (I can’t link directly because ASCO won’t provide a stable URL)]

Comment: Dr. Ratain's presentation complete with audio and slides is available on the ASCO website. There should be a link to it from the abstract page. The presentation is more detailed by far than the abstract.

This CancerGuide Page By Steve Dunn. © Steve Dunn
Page Created: October 10, 2003, Last Updated: June 22, 2004