Action to Cure Kidney Cancer

Tom Hutson, from Baylor-Sammons Cancer Center in Dallas, Texas, presented a poster at the 2008 ACOR conference updating the results of a Phase II study of pazopanib, a multi-targeted angiogenesis inhibitor (he first reported the interim results of the trial at ASCO 2007 - see our report from last year under Report from ASCO 2007). Like sunitinib (Sutent) or bevacizumab (Avastin), pazopanib works by curbing the growth of new blood vessels in tumor metastases. It does so by inhibiting vascular growth receptors - vascular endothelial growth factor receptor (VEGFR), platelet-derived endothelial growth factor (PEGFR) and c-Kit - and preventing the signal that would promote the growth of new blood vessels from reaching their target. This phase II study was designed to further measure the drug's efficacy and safety and, in addition, to examining the physiological changes that occurred in patients over the duration of the treatment and the genetic dispositions present in some patients that may influence the drug's activity.

Trial Design

225 RCC patients were administered pazopanib at 800 mg daily over a 12-week period, with RECIST-defined objective response being the trial's primary endpoint. Patients who responded to the drug were allowed to continue with it, and those who experienced stable disease during treatment were randomized to continue the drug or to be given a placebo, and their Progression-Free Survival was monitored. However, based on the positive activity of the drug, the randomization was discontinued, the patients on the placebo arm were crossed over to pazopanib, and the trial continued as a single-arm study.

Close to half the trial participants were rated as favorable risk and the other half intermediate risk with very few poor risk patients enrolled in the trial (risk ratings according to the Memorial Sloan-Kettering Cancer Center, MSKCC risk factors). 69% of the patients had prior therapy and the balance, 31%, were treatment-naïve.

Adverse Events

Of 225 patients, 215 (96%) experienced some degree of adverse event during the course of their treatment, with 34% having a grade 3 event and 7% having a grade 4 event. Two grade 5 (deadly) events were recorded - one instance of bowel perforation and one of severe dyspnea (difficulty in breathing). The most common side effects that accompanied pazopanib were: diarrhea (59%), hair color changes (43%), hypertension (40%), nausea (37%) and fatigue (37%), but most of these adverse events were grade 1/2 except for hypertension, which was grade 3 in 8% of patients.

The most common laboratory toxicities observed in patients on pazopanib were elevated liver enzymes (53%), lymphopenia (low level of lymphocytes) (45%), hyponatremia (low sodium level) (37%), leukopenia (low level of white blood cells) (35%) and creatinine elevation (32%). These toxic events usually didn't reach grade 3 or higher in severity, and the overall safety profile for pazopanib was considered manageable.

Results

Overall, pazopanib demonstrated quick and durable activity in patients, with a manageable side-effect burden. The results of the trial are compelling, with nearly 35% of patients responding to the drug and over 75% managing to avoid progressive disease. The full response rates are listed below:

Biomarker Analysis

The biomarker analysis portion of the pazopanib study served two purposes: one, to give researchers a sense of the physiological changes - the changes in blood chemicals, specifically those involved in signaling for vascular growth - that occurred in patients while on pazopanib, and two, to analyze how patients with mutations in the von Hippel-Lindau gene responded to the medication.

The blood biomarkers analyzed in patients during this trial were serum proteins CEC, VEGF (a pro-angiogenesis signaling peptide), soluble VEGFR1 (sVEGFR1) and soluble VEGFR2 (sVEGFR2). A baseline measurement for each was taken at the beginning of treatment, and samples were tested between 4-week intervals over the course of the pazopanib trial. No significant correlation was observed between changes in the levels of VEGF, VEGFR1 or CEC and solid tumor response. However, a decrease in serum levels of VEGFR2 (defined as a drop in concentration of 31% or more) was significantly correlated to tumor response and Progression-Free Survival. This observation demands more research (and a phase III trial is underway), but it could be very telling as to how pazopanib works physiologically and may offer insight into how to further refine its performance.

The von Hippel-Lindau protein is an important regulator of angiogenesis (blood vessel growth). Disruptions of the VHL gene - either destructive mutations or hyper-methylation (essentially, deactivation) - are common in RCC patients and in fact VHL is one of the few genes directly linked to kidney cancer. 40% of patients with a VHL mutation will develop RCC in their lifetime, on top of a host of other urological disorders[1]. Given this predisposition, it would be interesting to see how RCC patients with VHL mutations fare on medication relative to patients without, and if there is any correlation between drug response and genetic profile.

Dr. Hutson and his cohort were able to assess 78 of their 225 patients, and of that group found 70 (90%) with some form of VHL mutation or methylation. Among those patients, response rates were as follows: 41% had partial response, 40% had stable disease, 9% had progressive disease and 10% were not evaluable. By comparison, a group of 8 wild-type (non-mutated) patients from the sample had a different statistical performance. Among wild-type individuals, 13% had a complete response (one person), 25% had partial response, 38% had stable disease, 13% had progressive disease and 13% were not evaluable.

Discussion

Dr. Hutson's findings confirm that pazopanib has anti-tumor activity against RCC. The side effects of the drug seem to be manageable. The fact that 90% of the patients had a VHL mutation or methylation is understandable since the patients were all clear cell histology. However, Dr. Hutson did not find a correlation between VHL mutation/methylation and response to pazopanib therapy.

GlaxoSmithKline (GSK), the manufacturer of pazopanib, is running a large, 400-patient Phase 3 trial in Europe and elsewhere, testing pazopanib against a placebo in clear cell patients. The data collection date is April 2009, which should give the company time to present the results at ASCO 2009 and then apply for FDA approval for the drug. Efficacy in this trial may be good enough for the FDA to approve the drug, but in order to convince oncologists to prescribe it over the other targeted therapies already approved for renal cell, pazopanib will have to stand out among the increasingly crowded field. To this end, GSK is planning to run a head-to-head trial of pazopanib against sunitinib (Sutent) in a worldwide trial with over 800 participants at some 150 sites with Dr. Robert Motzer of Memorial Sloan-Kettering being the principal investigator. GSK will pay particular attention to monitoring heart function during the trial since there have been concerns that sunitinib treatment has caused adverse heart related events. GSK hopes to commence the trial in August 2008 (we learned of this in discussions with GSK reps at ASCO 2008 in Chicago in June).

[1] "Von Hippel-Lindau Syndrome; VHL" OMIM. http://www.ncbi.nlm.nih.gov/entrez/dispomim.cgi?id=193300.