Circumventing angiogenic escape mechanisms in kidney cancer
Much like Houdini, who freed himself from handcuffs and shackles, tumors find clever routes of escape from VEGF receptor inhibitors, a new class of drugs heralded for prolonging the survival of patients with renal cell carcinoma (RCC), the most common kidney cancer. In recent months, a team of DF/HCC investigators has been studying these angiogenic escape mechanisms and searching for new therapies to circumvent them.
Exploiting the biology of kidney cancer
Until a few years ago, metastatic RCC showed limited response to drug treatment, and less than 10 percent of patients with advanced disease survived to five years. However, the advent of inhibitors that block the vascular endothelial growth factor receptor (VEGF-R) has dramatically improved progression-free survival and overall clinical outcome for these patients. VEGF-R inhibitors work, explains William Kaelin, MD (DFCI) – whose laboratory laid the scientific foundation for their development – because kidney cancers are “notoriously rich in blood vessels” and therefore highly dependent on VEGF to maintain the tumor’s blood supply.
Although VEGF-R inhibitors, which include sunitinib and sorafenib, disrupt tumor angiogenesis and represent a major advance in kidney cancer treatment, they do not produce complete and durable responses, says Michael Atkins, MD (BIDMC), leader of the DF/HCC Kidney Cancer Program. “Within six to 12 months, the majority of patients develop resistance to these drugs, and there is currently no therapy that is effective in this situation.”
The physiology of resistance
The urgency of the problem led to a major project in the Kidney Cancer Program to study acquired resistance to VEGF-R blockade. Project leaders Atkins, S. Nahum Goldberg, MD, and James Mier, MD (all of BIDMC), draw on the resources of the Kidney Cancer SPORE (Specialized Program of Research Excellence) to identify the underlying mechanisms of resistance and to translate these insights into clinical trial approaches for preventing or delaying angiogenic escape.
The effort began with the development of mouse models of resistance. Goldberg’s lab created human RCC cell lines, implanted them into mice, and grew tumors to about 12mm. Investigators then treated the tumors with either sunitinib or sorafenib and monitored their size until the tumors began regrowing. At different time periods before and during therapy, they imaged the tumors using arterial spin labeling (ASL) MRI, a technique for measuring perfusion (the degree of vascularization). According to Goldberg, this method of imaging captures a far more precise picture of the physiology of the tumor than do traditional MRI scans.
“Areas of reperfusion represent zones of growing tumor that are detected even while other areas of tumor are shrinking [from treatment],” says Goldberg. “If you only measure how big the tumor is, you can’t tell what percentage is viable or what is actually happening inside the tumor. ASL-MRI enables one to determine the true impact of therapy and identify the onset and mechanism of resistance. It also provides an opportunity to biopsy specific areas of the tumor that are exhibiting resistance.”
Three key findings emerged from the xenograft models: (1) perfusion occurs before the tumor starts growing again – and may act as an early marker of tumor progression; (2) the appearance of perfusion on ASL-MRI mimics its histologic pattern; and (3) the more perfusion at baseline, the more likely the tumor is to respond to VEGF inhibitors. “We learned that when the tumor’s main mechanism of survival is blocked,” says Atkins, “it uses alternate mechanisms to reestablish perfusion.” Unlike resistance found in other cancers, such as gastrointestinal stromal tumors or lung cancer, he adds, these changes are not permanent and do not result from mutations in tumor cells but rather from physiologic changes that occur when the tumor is under stress from VEGF-R blockers. Indeed, when investigators reimplanted tumor cells into the animal models, the cells once again responded to VEGF-R blockade and later became resistant.
Profiling tumors: In search of new drug targets
Identifying the genes implicated in angiogenic escape is the contribution of James Mier, MD. Using Affimetrix gene arrays, he and colleagues compared the gene expression profiles of untreated, responsive, and resistant tumors from the xenograft models. They focused on those genes whose expression varied up or down more than two-fold at the RNA level in tumors that were regrowing despite treatment. Although numerous genes were differentially expressed, researchers were most interested in those known to be associated with a tumor’s ability to invade surrounding tissue or to recruit new blood vessels. The two dozen genes that met these criteria were then confirmed at the protein level.
Investigators identified three key classes of resistance-related genes: those that encode proteases (eg, MMP-1), immunosuppressive proteins (eg, arginase), and extracellular matrix proteins, which help support tumor growth. In addition, interferon-related genes declined in expression.
“Our main finding was a massive upregulation in protease genes, which are associated with increased invasiveness,” says Mier. Some of these proteases, he explains, are hypoxia-inducible factor (HIF)-dependent genes that any tissue will activate in response to the oxygen deprivation caused by VEGF-R blockade. Because many of the gene products are not druggable targets, Mier is also responsible for prioritizing the approaches that should be studied in animal models.
The goal is to move quickly into clinical trials, says Mier. “We’re focusing on the low-hanging fruit,” the proteins that are already targeted by FDA-approved drugs – including a Cox-2 inhibitor (which blocks arginase), interferon, and an MMP-1 inhibitor – which can be used in combination with sunitinib or sorafenib.
Translating insights into clinical trials
A team led by Atkins will soon begin clinical trials to confirm whether reperfusion, as shown on ASL-MRI, is a predictive marker of resistance in patients receiving sunitinib, and to intervene, once resistance is identified, with a drug that might block the angiogenic escape mechanism.
“By intervening with something as simple as celecoxib [a Cox-2 inhibitor], we might be able to interrupt the process,” says Atkins. The study will use imaging to determine whether the group receiving both sunitinib and celecoxib is doing better or worse relative to the group receiving only sunitinib or a placebo. Another approach, which could run in parallel, would intervene at the very beginning of treatment to see whether combining both drugs from the outset would further delay the onset of resistance.
Investigators hope that this research results in a more rational design of effective therapies for kidney cancer. From a broader perspective, says Atkins, since all tumors are dependent on angiogenesis, “working out some of the mechanisms of escape in kidney cancer may pave the way for similar studies in other cancers.”