Thinking therapeutically: Scientists in search of smarter tests and treatments for prostate cancer
Even though prostate cancer is the second leading cause of cancer death in men in the United States, some patients may wonder whether treatment is worse than the disease. The three primary therapies—prostatectomy, external-beam radiation, and brachytherapy—cause life-altering sexual, urinary, and rectal side effects that can linger for years. But what if a simple test could differentiate an indolent prostate cancer from a highly aggressive one?
“It would tell us who requires treatment and who can be monitored with active surveillance,” replies Philip Kantoff, MD (DFCI), leader of the DF/HCC Prostate Cancer Program and SPORE. Adds co-leader Lewis Cantley, PhD, deputy associate director for Basic Sciences at DF/HCC and director of the Cancer Center at BIDMC, “Such a test could save a huge amount of agony and probably a lot of unnecessary treatment.”
Thus disease stratification is one of the major themes of the Prostate Cancer Program. At the same time, investigators are searching for viable therapeutic targets in the pathways that fuel prostate tumors, while continuing their study of the genetic epidemiology of the disease (see previous CGEMS story).
Smarter mice may differentiate disease, predict clinical response
The critical question today, says Cantley, is whether a patient’s prostate cancer is slow-growing and localized or metastatic and deadly. “The simple tools we have now—needle biopsies and PSA measurements—do not allow us to discriminate.”
Cantley expects that his mouse models may help: revealing which mutations produce a lethal tumor and which result in a lethargic one, turning up some distinctive markers for new assays, and predicting which patients will respond to new therapies now being investigated.
These expectations may seem a bit lofty for mere mice. But those in Cantley’s lab are not the typical xenografts, immune-deficient rodents into which human cancer cells have been transplanted; such models have proven to be poor predictors of clinical response. Cantley’s mouse models are genetically engineered to develop prostate tumors over time within a normal microenvironment with an intact immune system.
Each model replicates a single genetic event implicated in human prostate cancer, the most common of which are the loss of PTEN, p53, and p27; a rearrangement of the fusion gene TMPRSS2:ERG (a research area of Todd Golub, MD, Levi Garraway, MD, PhD, and William Hahn, MD, PhD of DFCI); and activating mutations in PI3K. Researchers then cross the single-gene models to produce mice that mimic the genetic combinations causing early and late-stage human disease.
As the mice develop tumors, they are treated with inhibitors, now in phase I clinical trials, that target members of the PI3K pathway, including PI3K, PTEN, and mTOR. Since PTEN is the brake to PI3K’s accelerator, inhibitors of PI3K can theoretically decelerate the growth of tumor cells missing PTEN; Pier Paolo Pandolfi, MD, PhD (BIDMC), a world-renowned animal modeler, has developed mouse models with prostate-specific deletions of PTEN that will be used in these studies. Moreover, since recent research by Thomas Roberts, PhD (DFCI) and Jean Zhao, PhD (HMS) showed that both the p110-alpha and p110-beta isoforms of PI3K must be inhibited to block tumors specifically driven by the loss of PTEN, animal models are treated with inhibitors of both isoforms.
Cantley hopes that new therapies, like these inhibitors, can one day treat localized tumors and reduce the need for surgery, prevent micrometastases from growing into full-blown tumors, and save lives in end-stage disease, for which there is currently no cure. PI3K inhibitors might also be good adjuvant therapies for halting micrometastases—independent of whether PI3K is driving the tumor or not, says Cantley—because these agents interfere with developing vasculature, which small tumors rely on to grow.
Investigators will collect and analyze patient tumor, correlate the mutations with patient outcomes, and compare the results with their observations in mice. “We have a wealth of PI3K inhibitors, but none has been tested yet in prostate cancer. So as these agents are being evaluated for safety in humans, we’ll be testing the same drugs in our mice and asking: ‘In which mutational events do these drugs work?’” The answers will guide investigators in the design of phase II clinical trials by identifying which patients are most likely to respond. “We’re very optimistic that inhibitors of the PI3K pathway will be effective in at least a subset of prostate cancers,” Cantley predicts, “though they will probably have to be combined with other drugs to have a big impact.”
Seeking new targets in androgen-independent prostate cancer
Taking an alternate route to the same goal, Myles Brown, MD, chief of the Division of Molecular and Cellular Oncology at DFCI, is studying the androgen receptor pathway to uncover new therapeutic targets in metastatic disease. Once prostate cancer metastasizes (typically to bone) and is no longer treatable by surgery or radiation, first-line treatment remains androgen-deprivation therapy (ADT). ADT blocks the production of circulating androgens, on which all tumors initially depend, and shrinks them at first. But they eventually lose their dependence on androgens and grow back in metastatic sites, explains Brown. He believes that understanding what controls the growth of androgen-independent prostate cancer may lead to new treatment strategies.
Like other steroid hormones, androgens work by binding to intracellular receptors and acting as transcription factors to regulate a specific gene program. Tumors that become androgen-independent have discovered clever ways to sustain themselves despite a lack of hormones: some make their own androgens (the research focus of Steven Balk, MD, PhD of BIDMC), while others appear to acquire androgen receptors (AR) that exhibit ligand-independent transcriptional activity. Brown has been studying the latter in search of a smarter class of AR antagonists that block the receptor’s ability to turn on genes.
“Interestingly, in the system we’ve been studying, it looks as if AR is executing a different gene program than it does in androgen-dependent prostate cancer, and is turning on genes that regulate growth in another way,” says Brown. Using tumor arrays from Balk and Max Loda, MD (DFCI), his lab has validated these genes as the same targets upregulated in clinical cases of androgen-independent prostate cancer. “Our hope is that some of these newly turned on genes become targets of therapy,” says Brown, “because this type of prostate cancer is a major area of unmet need. It’s what men die of.”
Treating long-term complications of hormone therapy
The men who survive prostate cancer now number about two million in the United States. Of these, at least one-third are being treated with hormone therapy at any given time, says physician-scientist Matthew Smith, MD, PhD, director of Genitourinary Medical Oncology at MGH. While hot flashes and loss of sexual desire have long been recognized as side effects of hormone therapy, says Smith, it was not until the PSA test led to earlier diagnosis and longer treatment that he and others began to witness another unexpected side effect: osteoporosis. Smith, who believes that estrogen deficiency explains this complication, has designed major clinical trials to test new treatments for men suffering this side effect.
Estrogen, a key regulator of bone metabolism, may at first seem an unlikely culprit in a male-only cancer. But as Smith explains, older men normally have higher levels of estrogen than do postmenopausal women, whose estrogen levels have dropped. Since testosterone metabolizes to estrogen in men, and hormone therapies dramatically deplete testosterone, men’s levels of estrogen also fall—leading to higher bone turnover, net bone loss, and a greater risk of fractures.
Smith and colleagues were among the first to recognize the long-term consequences of hormone therapy and to lead efforts to prevent treatment-related osteoporosis. One placebo-controlled study of 1,400 men evaluated the efficacy of denosumab, a monoclonal antibody, in treating bone loss from hormone therapy; it demonstrated a significant increase in bone mineral density and decrease in fracture risk, reports Smith. A second trial tested toremifene, a selective estrogen receptor modulator, with similar results.
“These phase III studies arose from our research that helped define the problem and explain its underlying mechanism,” he says. “Most of what we knew before was inferred from postmenopausal women. But we wanted to know whether these interventions would work in this specific patient population; these studies have shown we can prevent fractures with these drugs.” Smith is now leading a global clinical trial to see whether denosumab can prevent bone metastases in men with high-risk prostate cancer.
The research of Cantley, Brown, Smith, and many others illustrates that the DF/HCC Prostate Cancer Program is moving in the right direction, says Kantoff. “Over the last five years, we’ve been thinking more therapeutically to translate basic science discoveries to the clinic. But our therapeutic portfolio is not complete yet,” he adds. “I’m hoping that over the next five to ten years, with the great wealth of scientific talent at DF/HCC, we’ll have a startling impact on survival and quality of life.”