Special Report: Cancer
An ever-evolving picture
Cancer heterogeneity can make therapy feel like whack-a-mole
By Randall C Willis
It’s summer and your family has decided to take a cottage vacation. Without television, internet or cell service, the nights are quiet; too quiet for the kids. Scrounging for activities, you find an old jigsaw puzzle and hope that all the pieces are still there.
At first, the kids don’t buy into it; but after assembling one edge and a chunk down the middle, they become more engrossed in the activity. And just as everyone finally goes for broke, something odd happens: the picture changes.
Pieces that once fit together, no longer do, and in frustration, you pull sections of the puzzle apart to reassemble them a different way…only for the photo to change again…and again…and again.
Sure, the family is occupied, but for how long? How many times can they restructure the puzzle until they finally give up?
Unfortunately, for a lot of cancer patients and the oncology teams treating them, giving up isn’t really an option. Giving up on the puzzle potentially means giving up on life itself, because the constantly morphing puzzle is the cancer itself.
Malignancies that once responded to treatment no longer do, whether because of acquired drug resistance mutations or because drug-tolerant or drug-resistant cells that once formed the minority of the cell population have become predominant.
Understanding change
“Tumor heterogeneity contributes to the adaptive potential of tumors and plays key roles in resistance to treatment,” offers Derek Ostertag, director of R&D Diagnostics at Tocagen. “The more heterogeneous a tumor is, the more rapidly it will likely acquire treatment resistance.”
And it is this evolving resistance that continually thwarts efforts to treat patients with even the most targeted of therapies.
George Karlin-Neumann, scientific affairs director for Bio-Rad’s Digital Biology Center, recalls the initial excitement over the impact of vemurafenib, targeting BRAFV600, on melanoma, describing the rapid and widespread disappearance of skin nodules—the difference between the before and after panels—as magic.
“But what wasn’t on the Nature cover was the third panel, which we now see at conferences, which was that [the nodules] came back—and at the same sites, too,” he says ruefully.
“I think we always knew that [heterogeneity] was a problem because basically, it is almost a rule that every patient who gets a partial remission of their cancer ultimately progresses,” adds Stephen Marcus, CEO of Cantex Pharmaceuticals. “Something happens that changes the tumor and makes it resistant to therapy.”
“With the targeted therapies, people were optimistic that if you had a great target and you hit the target, then the cell would die,” Marcus continues. “It turns out that the cell evolves and evades those targeted therapies.”
Adding to the complexity of tumor heterogeneity is that genetic mutations are not the only intrinsic factors that may change. Rather, according to Xiao-Xiao Sun and Qiang Yu of the Shanghai Institute of Materia Medica, epigenetic changes may also have significant impact.
“Studies on [cancer stem cells; CSCs] suggest that a small number of cells in a tumor undergo epigenetic changes, similar to the differentiation of normal stem cells, to form phenotypically diverse non-tumorigenic cells that compose the bulk of the cells in a tumor,” the researchers wrote in Acta Pharmacologica Sinica last year. “Direct evidence of differences in epigenetic changes between tumor subpopulations have also emerged.”
Similarly, they suggested, changes in methylation patterns can be quite chaotic and constantly evolving throughout tumorigenesis, having significant impacts on gene expression patterns from cell to cell and over time. Such expression differences can lead to alterations in cell fate between sister cells and in the responses of signal transduction cascades to environmental factors.
As Ostertag points out, however, the source of heterogeneity may not be restricted to the tumor itself.
“Our understanding has expanded to recognize that tumor heterogeneity refers not only to the genetic diversity that exists in each patient’s tumor, but also to the diversity in the tumor microenvironment,” he explains. “This can include physical barriers—e.g., extracellular matrix (dis)organization, necrosis, hypoxia, fibrotic tissue—a varied supporting cast of non-cancerous cells that have evolved with the tumor, varying degrees of infiltrating immune cells and diversity in the tumor microbiome.”
Thus, the oncology researchers and clinicians are met with a cascade of factors than can complicate not only decisions on potential therapeutics and possible treatment regimens, but also the fundamental prognosis of the patient.
In January, Luc Morris and colleagues at Memorial Sloan Kettering Cancer Center published their efforts to identify correlations between intratumor heterogeneity (ITH) and overall survival. Publishing in Oncotarget, the researchers analyzed clinical and genetic data from almost 3,400 tumors across nine cancer types.
“Genetically heterogeneous tumors, comprised of multiple subclonal populations, tend to be associated with poorer patient survival than tumors harboring low/moderate levels of intratumor heterogeneity,” the researchers noted. “We found that the prognostic value of ITH was significant in many cancer types, and in multivariable analyses, remained significant when controlling for other relevant clinical, pathologic and molecular features.”
They also noted an inverse correlation between ITH and immune cell infiltration across all cancer types, although immune infiltration did not appear to mediate the poorer survival trend.
Adding yet another layer of complexity is the challenge that heterogeneity can occur between tumors within a single patient.
In February, Ryan Corcoran of Massachusetts General Hospital Cancer Center and colleagues highlighted this challenge, monitoring drug resistance acquisition in colorectal cancer metastases with radiographic imaging as well as next-generation sequencing and Droplet Digital PCR (ddPCR) analysis of tissue and liquid biopsies.
As they described in Cancer Discovery, when a tissue biopsy suggested a MEK1 mutation in a single progressing liver metastasis, the clinicians switched the patient from treatment with cetuximab to a combination of panitumumab and trametinib. As hoped, imaging revealed that the lesion harboring the MEK1 mutation responded, corroborated by ddPCR analysis of circulating tumor DNA (ctDNA).
Unexpectedly, however, another metastasis progressed during treatment and was shown to carry a KRAS resistance mechanism, something missed by the original tissue biopsy. The secondary resistance pathway highlights the challenge of lesion-specific response to targeted therapy.
“Our original single-lesion biopsy was not sufficient to capture the molecular heterogeneity of this patient’s cancer and failed to detect the simultaneous presence of an additional resistance mechanism (KRAS mutation) that eventually led to treatment failure,” the researchers wrote. “However, both mutations were readily detectable in ctDNA from blood collected prior to combinatorial therapy.”
“Not only did real-time ctDNA analysis enable identification of a second resistance mechanism not captured by the single-lesion biopsy, but it did so while the patient still appeared to be responding to therapy, thereby predicting both the timing and cause of impending treatment failure,” they continued.
Despite the success in this example, however, the measurement and identification of ctDNA (or cell-free DNA; cfDNA) isn’t without its own susceptibility to heterogeneity, as even a single cancer can behave differently when examined at different time points or stages of development. This point was demonstrated earlier this year by Geoffrey Oxnard and colleagues at Dana-Farber Cancer Institute in JAMA Oncology.
The researchers used ddPCR to genotype cfDNA for EGFR and KRAS mutations in patients with newly diagnosed advanced non-small cell lung cancer (NSCLC) and those with acquired resistance to EGFR kinase inhibitors. Primarily, they wanted to determine the feasibility and accuracy of such assays, but they also wanted to determine the diagnostic turn-around time vs. traditional tissue genotyping.
Although the plasma ddPCR assay was highly predictive for all of the mutations tested, the sensitivity of the assay was lower for some mutations than others. What was more interesting, however, was that the sensitivity increased significantly in patients with hepatic and bone metastases and in those with increased number of metastatic sites.
“This newly demonstrated relationship is likely related to increased cfDNA shed in the setting of more extensive disease where tumor cfDNA shed is the chief driver of assay sensitivity and determines its upper limit,” the authors wrote.
Perhaps just as importantly, the researchers were able to identify plasma genotypes within two to three days of a blood draw, a dramatic improvement over the turn-around time for tissue genotyping of 12 (newly diagnosed) or 27 (resistance) days.
Thus, even with reduced sensitivity, the ability to know that information sooner may prove helpful.
“If you can get half the patients and figure out what to do and do it more quickly, it’s beneficial for them,” opines Karlin-Neumann. “For the other ones, I’m sorry that I can’t help you as quickly as I might like to, but we can still do something else [e.g., tissue biopsy] that might tell us what to do.”
Modeling chaos
Given this molecular maelstrom, much effort has gone into trying to mimic the heterogeneity in the lab setting to hopefully understand it better, as well as develop better assays for therapeutic screens. And as Megan MacBride, associate director of product management at Taconic Biosciences, explains, modeling this complexity has meant an expansion from simple cell cultures.
“Cell lines are very simple tools, very homogenous,” she says. “It really gives you a go/no-go, but it doesn’t give you the clinical relevance.”
To some extent, the growing interest in immunotherapies has led to a renaissance of animal models as researchers look to study the impact of therapeutics within the context of a fully functioning immune system. Thus, companies like Charles River Laboratories and Crown Bioscience have strengthened their efforts to develop allograft mouse tumor systems—also known as syngeneic/syngenic models—for any number of human cancers (see also “Crown makes oncology, cardiovascular and metabolic strides” in the May 2016 issue of DDNews).
Despite these efforts, however, there is an open question as to whether animal models provide only limited insights into the diseases and their treatment, and the refrain of curing cancer in mice (but not humans) is repeated yet again.
“Advances in tissue and cell type specificity have demonstrated the importance of mouse models for studying intertumor and intratumor heterogeneity of breast cancers,” suggested Shany Koren and Mohamed Bentires-Alj of the Friedrich Miescher Institute for Biomedical Research. “But the question remains how accurately murine models reflect human breast cancer heterogeneity.”
“Histological phenotypes of several mouse mammary tumors do not resemble human breast cancers,” they wrote in Molecular Cell, “and the frequency of hormone-dependent mammary cancers is much lower in mouse models than in humans.”
This is where the advent of patient-derived xenografts (PDXs), involving the transfer of primary human tumors from patients into immunodeficient mice, has helped, says MacBride, allowing researchers to replicate the heterogeneity of human cancers.
“There are a range of animal models that go from nudes, which just lack T cells, all the way up to what we call the NOG mouse, which is a super-immunodeficient that doesn’t have T cells, B cells or NK [natural killer] cells, and what’s left is impaired for various reasons,” MacBride explains. “It has macrophages, it has dendritic cells, but they don’t work like in a normal mouse.”
At the same time, Koren and Bentires-Alj noted a significant limitation of even these systems.
“Because of cross-species incompatibilities and their transplantation into immunodeficient mice, human cell line xenografts and PDX models both lack the contribution of some non-cell autonomous drivers and the immune system to tumor heterogeneity,” they wrote.
Acknowledging this challenge, companies like Taconic have worked to renew the immunocompetence of their model systems by recreating human immune systems within their mice.
In the huNOG line, for example, human hematopoetic stem cells have been introduced into NOG mice to essentially generate a functional T cell model.
“And that’s great if you’re looking at an immune checkpoint inhibitor,” MacBride says. “If you want to study Keytruda or Yervoy, T cells are typically enough.”
Because some of the mouse cytokines cross-react during cell differentiation, the development of a lymphoid system is great, she continues, but myeloid development is not. Thus, even though the immune system has been humanized, it is still not fully representative of the broader human repertoire.
To address this issue, the company has also created models that replace the non-cross-reactive mouse cytokines with their human counterparts.
The huNOG-EXL mouse, for example, expresses human GM-CSF and IL3 that push the stem cells down the myeloid pathway. Similarly, hIL2-NOG and hIL6-NOG support human NK cell engraftment and human monocyte/macrophage engraftment, respectively.
And of course, Taconic is not alone in this pursuit.
At the recent American Association for Cancer Research (AACR) conference in New Orleans, The Jackson Laboratory presented their efforts to evaluate immune checkpoint inhibitors in their Onco-Hu mouse models. When engrafted with human hematopoetic stem cells, immunodeficient NSG mice can support the development of many components of the human immune system. The related NSG-SGM3 lines, meanwhile, also express SCF, GM-CSF and IL3.
In NSG mice bearing PD-L1-positive NSCLC PDXs, treatment with pembrolizumab (anti-PD-1-receptor) resulted in tumor growth inhibition as well as increased CD8+ T cell infiltration. The same was true when NSG-SGM3 mice engrafted with triple-negative breast cancer PDXs were treated with ipilimumab (anti-CTLA-4).
“Our new humanized mice enable researchers studying immuno-oncology to obtain results from mouse studies that may be more predictive of therapeutic response in humans,” said Walter Ausserer, business director for Clinical & In Vivo Services, in the announcement. “These models represent an important step toward modulating the immune system to recognize and attack tumors.”
Defining diversity
Of course, even when you know a lot about the specific markers that help characterize and define cancer heterogeneity, being able to identify and monitor those markers in a given patient can still be a significant challenge. Perform a tissue biopsy of even a single cancer lesion from any of the three Cartesian directions and you may believe you’re looking at three different tumors with markers that are shared by all, some and none of the biopsies.
This is one reason why liquid biopsies are becoming more prevalent in cancer research, as was obvious on the exhibition floor of the recent AACR conference. As suggested above (and in “Non-invasion of the body snatchers” in the October 2015 issue of DDNews), by sampling body fluids such as blood, urine or even cerebrospinal fluid, researchers can probe circulating tumor cells (CTCs), cfDNA and even extracellular vesicles (EVs) for changes in mutational signatures or gene-expression patterns.
But as noted by the Dana-Farber researchers, one analytical method may not be sufficient to give a complete picture.
“It is critical to have a strong understanding of the hypothesis being tested—and what type of sampling you will need—to have high confidence that you have adequately characterized a patient’s tumor with the data collected,” argues Ostertag. “This complexity is one of the reasons why validated biomarkers in cancer settings have been slow to emerge.”
Thus, some companies are looking at a broader platform of resources and services to provide more comprehensive information.
Combining the worlds of liquid biopsy and tissue screening, for example, Helomics recently launched two personalized healthcare systems—Defender and CellFx—as part of its broader precision cellular analytical platform, or PCAP. CellFx allows Helomics staff to grow cultures from individual patient’s tumors and screen them for gene and protein expression patterns associated with cell proliferation, cell death, drug response and cytotoxicity over time.
The goal is to ultimately offer clinicians insights on the biomolecular nature of a given patient’s disease that should lead to better-informed treatment decisions. Similarly, the improved understanding should also help researchers identify and develop new therapeutic approaches and potential biomarkers, whether prognostic, diagnostic or therapeutic.
Specialists in nuclear medicine, Advanced Accelerator Applications (AAA) approaches heterogeneity analysis from a molecular imaging perspective, relying on many of the basic radiological platforms that have been used in hospitals for years.
“In the case of [neuroendocrine tumors; NETs], for example, you might use three different diagnostic in-vivo examinations,” explains Stefano Buono, AAA’s CEO. “You might use PET with our Somakit-TATE for somatostatin expression, FDG (18F-fludeoxyglucose) for proliferation rates and you could even use F-DOPA.” [Editor's note: Since the interview was conducted with Buono and the article submitted, Somakit-TATE has been approved and is now named NETSPOT]
Detecting targets with different tracers, he suggests, might be a strong signal of heterogeneity of the tumor. So, for the full analysis of a patient, perhaps more than one diagnostic tool has to be used to determine the heterogeneity.
Independently echoing Buono, Bio-Rad’s Karlin-Neumann hearkens back to the lessons learned at Dana-Farber, making reference to clinicians with whom he has working relationships.
“They don’t see tissue biopsies going away,” he says. “It’s not like we lived in the Dark Ages and now we’re enlightened; let’s just do cfDNA. It was very enlightening to me as to how they’re taking the technologies that have emerged and asking how good are they and how well do they match up with the biology, which we’re just learning about.”
And there still remains the logistical issues behind sampling, says Ostertag. Complex sampling is difficult with multicenter clinical trials, and often the best scientific approach to sample collection is not feasible in large multicenter trials.
“For example, reliably procuring properly handled tumor tissue is an enormous challenge in statistical validation of protocols which can require large numbers of patients from multiple sites,” he explains. “For this reason alone, blood samples are generally a preferred sample type to collect.”
Possibly simplifying this challenge, however, are the results of a recent study by Washington University School of Medicine’s Timothy Levy and colleagues, published last year in JAMA.
The researchers performed whole genome sequencing on paired bone marrow samples from 50 AML patients taken at diagnosis and approximately 30 days after successful induction therapy when patients achieved remission.
The researchers found that if a patient achieved complete clearance of leukemia-associated mutations by day 30, their median time for event-free survival was almost triple that of patients who had even one mutation remaining (17.9 months vs. 6 months). Similarly, there was a fourfold improvement in terms of overall survival (42.2 months vs. 10.5 months).
At the AACR conference, Levy suggested that the question of outcome was less about what specific mutations carried through treatment than the idea that mutations carried through treatment. From his perspective, the findings opened the door to what he termed a “mutation-agnostic” approach to treatment.
One-two (three?) punch
In an era predicated on targeted therapeutic approaches, might a mutation-agnostic approach pump new life into old chemotherapies?
“In a sense, we are going back to the future by relying on chemotherapeutics with a better understanding of how to target available therapies,” says Ostertag. “I don’t think it will be focused on particular pathways.”
Mutations that help a tumor survive, whether driver mutations or other essential mutations, are readily being identified, he says. And that information on conserved pathways could make targeted therapy more effective.
“However, similar to chemotherapeutics, it is unlikely that a specific pathway will prove to be a universally effective target for a particular cancer indication,” he cautions. “Instead, identification of these pathways is expected to become part of the characterization of an individual patient’s tumor.”
Instead of focusing on individual pathways and targets, clinicians may want to think about broader and more fundamental therapies, perhaps used in combination with treatments targeted at significant cancer drivers on a patient-by-patient basis.
Tocagen’s lead product, for example, combines the retroviral replicating vector (RRV) Toca 511 with Toca FC, a prodrug derivative of 5-fluorocytosine.
“Toca 511 selectively infects tumor tissue rather than normal tissue,” Ostertag explains. “The virus spreads without killing infected tumor, and inserts a gene into the tumor genome that produces cytosine deaminase.”
The patient is then given Toca FC, which is converted by cytosine deaminase into the antimetabolite 5-fluorouracil, which kills tumor and local immune suppressive cells (e.g., myeloid-derived suppressor cells; MDSCs). Cancer cell death then triggers an immune response against the tumor.
“The combination stimulates broad but tumor-specific immune responses against an individual’s own tumor and thus has the potential to overcome much of the tumor heterogeneity issue,” Ostertag says, although he admits that the treatment’s effectiveness may yet depend on some aspects of heterogeneity.
“For example, we know that the virus spreads at different rates in different individual cancer cells and tumors in animals, and clinical outcomes vary,” he concedes. “That being said, the inherent multiple mechanisms of anti-tumor activity provided by Toca 511 and Toca FC therapy, which target many aspects of tumor biology, make it more difficult for even heterogeneous tumors to evade treatment.”
Cantex Pharmaceuticals, meanwhile, is pushing two projects through its pipeline that may function by hitting multiple targets simultaneously.
According to Marcus, CX-01 is a heparin derivative that showed activity in inflammatory lung disease as well as pancreatic cancer. The drug is being tested currently against acute myeloid leukemia, where it is believed to alter a pathway known as the CXCL12-CXCL4 axis that helps tether leukemia cells to the protective environment of the bone marrow.
“What we thought was we drove those cells out of the marrow into peripheral circulation where they were killed by the chemotherapy,” he explains. “When the chemo stops, the drug stops and its pharmacologic effect stops, allowing the normal hematopoetic stem cells that survive the chemotherapy to traffic back to the marrow and repopulate it.”
The drug has also demonstrated effectiveness through another pathway in facilitating platelet recovery, and therefore potentially offers some protection against thrombocytopenia.
The company is also repurposing the old alcohol-aversion therapy disulfiram into a treatment for recurrent glioblastoma based on its long anecdotal history (see “A story about anecdotes” in the May 2016 issue of DDNews) of anticancer activity and early University of Utah studies in this area.
“What’s interesting is that the target is actually uncertain,” Marcus says. “We know of a number of things that it does. It’s a proteasome inhibitor. It also may sensitize tumors to alkylating agents. The mechanism hasn’t been fully delineated, but it clearly has activity.”
Thus, it may be that by overwhelming cancer cells from multiple angles, disulfiram is able to circumvent any mutational advantages certain cells might otherwise experience.
For its part, AAA is leveraging its expertise in nuclear diagnostics into targeted radiotherapeutics. Its lead candidate for treatment of gastro-entero-pancreatic NETs is Lutathera, a 117Lu-labeled somatostatin peptide analogue.
According to Buono, although somatostatin receptors are relatively ubiquitous, they are highly overexpressed in NETs, making them attractive targets for Lutathera. But what makes this approach particularly successful is the fact that when the drug binds to a target cell, it doesn’t just kill that cell, but also cells in the immediate vicinity, whether they overexpress the receptor or not.
“The cytotoxic effect is the radiation coming from electrons that are emitted by the drug,” he explains, suggesting the electrons can travel 2 mm to 3 mm into the surrounding tissues.
“When you reach a target, with the same electrons you may irradiate a few millimeters around the origin of the radiation,” he continues. “So, when you treat a tumor, you get a crossfire effect within the tumor.”
By taking out a swath of surrounding cells, the company effectively cuts the immediate heterogeneity concerns off at the knees.
Working with Martin Pomper and colleagues at Johns Hopkins University, AAA is also developing a similar platform for prostate cancer, relying on the overexpression of the prostate-specific membrane antigen (PSMA), specifically PSMA-SR6. The goal is to develop not only the radiotherapeutic 177Lu-PSMA-SR6, but also the companion diagnostic 68Ga-PSMA-SR6.
These successes aside, coping with this complexity can be daunting to anyone involved in cancer research and treatment, and the seemingly uphill battle with the odds stacked against you can leave you depressed.
“I think that can make you a little pessimistic about whether the way we’re going about it is ever really going to get so much better,” says Karlin-Neumann. “It still feels a little like whack-a-mole.”
Accepting the inevitable
The irony of truly beginning to understand the breadth and depth of the heterogeneity challenge in cancer, however, is that it may just show us that we are helpless to change it—at best, we can merely define and monitor it, and we are getting better at that. But, by accepting this helplessness, we might just gain the freedom to focus on the things that we can change.
And make the puzzle easier to put together.
ADDITIONAL NEWS FROM THE WORLD OF CANCER R&D
Autism and cancer share many risk genes in common
SACRAMENTO, Calif.—Autism and cancer share more than 40 risk genes, suggesting that common mechanisms underlying the functions of some of these genes could conceivably be leveraged to develop therapies not just for cancer but for autism as well, an extensive assessment by researchers with the MIND Institute and the Comprehensive Cancer Center at University of California, Davis (UC Davis) has found.
“This striking coincidence of a remarkably large number of genes implicated in both autism spectrum disorder (ASD) and cancers has not been previously highlighted in the scientific literature,” said Jacqueline Crawley, MIND Institute distinguished professor and endowed chair, who added that the research identified 43 specific genes with autism susceptibility that also have an association with cancer. “Potentially common biological mechanisms suggest that it may be possible to repurpose drug treatments for cancer as potential therapeutics for neurodevelopmental disorders.”
“Autism and Cancer Shared Risk Genes, Pathways and Drug Targets” was published online in Trends in Genetics. Crawley collaborated on the work with professor and chair of the UC Davis Department of Microbiology and Molecular Genetics Wolf-Dietrich Heyer, who is affiliated with the Comprehensive Cancer Center, and Janine LaSalle, professor of medical microbiology and immunology, who is associated with the MIND Institute.
“Autisms” are best conceptualized like cancers, the authors write—in the plural. Like cancers, the behaviorally defined condition encompasses a broad range of putative causes, symptoms and outcomes.
Included in the dozens of genes implicated in both cancer and autism are genes for relatively rare syndromes, such as Rett syndrome and tuberous sclerosis, whose sufferers experience an array of physical and neurological symptoms, including intellectual disability, as well as communication deficits characterized as autism.
So what does tumor cell proliferation have in common with synapse formation and brain development?
“Errors associated with genome maintenance during fetal life may occur at critical time periods for [brain development] resulting in neurodevelopmental disorders,” said Heyer, “whereas errors more commonly occur during adult life in cell types susceptible to tumors.”
“Genes encoding nuclear proteins involved in epigenetic functions were frequently shared between cancer and ASD, implicating the importance of aberrant gene regulation in both disease states,” LaSalle added.
Considerable translational value can be gained from a new focus on understanding the genetic commonalities of autisms and cancers, they maintain.
“It may be possible to repurpose available cancer drugs with reasonable safety profiles as targeted treatments for ASD,” the authors write. “Stratifying individuals with ASD who harbor a risk gene for autism that is also a risk gene for cancer may enable therapeutic development of personalized medicines based on the specific causal mutation.”
High levels of protein p62 predict liver cancer recurrence
Study in mice and human tissues suggest new prognostic and therapeutic approach
SAN DIEGO—Researchers at the University of California, San Diego’s medical school and Sanford Burnham Prebys Medical Discovery Institute have discovered that high levels of the protein p62 in human liver samples are strongly associated with cancer recurrence and reduced patient survival. In mice, they also found that p62 is required for liver cancer to form.
The study, published May 19 in Cancer Cell, suggests p62 could be used as a prognostic marker and potential therapeutic target for liver cancer.
“By defining factors that allow liver cells to progress from pre-cancer to cancer, we were able to find one—p62—that we can also use to predict a liver cancer patient’s outcome following full removal of a previous liver tumor,” said co-senior author Dr. Michael Karin, Distinguished Professor of Pharmacology and Pathology and Ben and Wanda Hildyard Chair for Mitochondrial and Metabolic Diseases at the UC San Diego School of Medicine.
Karin led the study with co-senior author Dr. Jorge Moscat, deputy director of the Cancer Center at Sanford Burnham Prebys Medical Discovery Institute, and first author Dr. Atsushi Umemura, a postdoctoral fellow in Karin’s lab.
Protein p62 normally acts as the cell’s trash collector, delivering specially tagged proteins to the cell’s degradation machinery. P62 also acts as a communication hub—it binds many different proteins to regulate important cellular function like growth and survival. Amounts of p62 are known to be elevated in many different cancers, including liver, and in pre-cancerous liver diseases.
In this study, Karin’s team looked at non-cancerous liver samples collected from people who had undergone previous treatment to completely destroy their liver cancers. They graded the livers from zero to 3 based on the average number of p62-positive aggregates detected. Seventy-nine of 121 specimens were p62 positive. Using the medical records corresponding to each liver sample, the team also noted the number of years each patient survived disease-free.
The researchers found that people with high-grade p62 were significantly more likely to see their cancer return and less likely to survive cancer-free than people with low or no p62. They found the same correlation when they looked at the link between the p62 gene and survival outcomes for an additional 450 liver cancer patients whose genomic data and clinical records are available in national research databases.
Work in mice led the researchers to attribute protein p62’s pro-cancer effect to its ability to activate other proteins (NRF2, mTORC1 and c-Myc) and genes that help stressed cells survive. This extended lifespan allows liver cells to accumulate cancer-causing mutations and ultimately form malignant tumors. The researchers found that p62 alone was enough to induce liver cancer in several mouse models of the disease. Liver tumors couldn’t form without the protein.
The specific type of liver cancer analyzed in this study was hepatocellular carcinoma, the most common form of adult liver cancer.
“Our new study illustrates that p62 is necessary and sufficient to induce liver cancer in mice, and that its high expression level in liver tissue surrounding a tumor predicts recurrence of the disease after tumors are removed,” said Moscat. “We believe that small molecules that interfere with p62 may be useful for preventing the progression of chronic liver disease to liver cancer.”
Cancer clinical trials to be streamlined to develop new treatments sooner
LONDON—Set-up and delivery of multicenter cancer clinical trials in the United Kingdom will become more efficient and easier to get off the ground, thanks to a new agreement between U.K. cancer clinical trial centers.
The Experimental Cancer Medicines Centre (ECMC) Collaboration Agreement is expected to improve the network’s ability to carry out early-phase clinical cancer trials, boosting research and development in the United Kingdom and bringing innovative new treatments to patients sooner. The ECMC Network is a joint initiative between Cancer Research UK and the four health departments of the United Kingdom. The centers help the pharmaceutical industry and academic funders develop cancer drugs in early-phase clinical trials by bringing together teams of world-leading clinical experts.
Clinical trials in the United Kingdom often face delays because of the variation in how they are set up in each center. But under the new agreement, all 18 ECMC locations will do this in the same way and will work to the same standards—cutting paperwork and helping researchers work more collaboratively.
This will help speed up approval of early-phase clinical trials, making the United Kingdom a more attractive location for international pharmaceutical and life-sciences companies, those behind the effort say. It also means U.K. patients will get access to innovative treatments sooner.
“It’s essential for the U.K. to deliver clinical trials more efficiently if it is to build its reputation as a world-leader in early-phase clinical research,” said Aoife Regan, head of the ECMC Network. “We already have some of the best early-phase clinical trial researchers in the world, and this agreement builds on that by setting new standards for the set-up and delivery of clinical trials. By getting innovative treatments to cancer patients more quickly, we hope to build our international reputation and increase our ability to attract more companies to the U.K.”
George Freeman, a member of Parliament and life-sciences minister, added: “Effective clinical trials ensure that patients can benefit from the world-leading medical research taking place across the country. The Experimental Cancer Medicine Centre Network shows how collaborative work between charities and the government is driving forward new cancer discoveries.”
“We’re delighted that our ECMC Network is continuing to help cancer patients around the country benefit from world-leading research taking place here in the U.K.,” concluded Sir Harpal Kumar, Cancer Research UK’s chief executive. “This initiative will ultimately help potentially life-saving drugs reach cancer patients sooner by accelerating the first step of clinical research.”
The ECMC Network, launched in 2007, is an initiative funded in partnership between Cancer Research UK and the health departments of England, Scotland, Northern Ireland and Wales. Cancer Research UK is one of the world’s leading cancer charities, and receives no government funding as it pursues pioneering work into the prevention, diagnosis and treatment of cancer.
Cancer roundup
A look at some recent news on the oncology R&D front
By Jeffrey Bouley
BRUSSELS, Belgium—Biotech company argenx announced in April that it had signed what it calls “a major deal with AbbVie in the field of immuno-oncology.” The agreement is the result of a collaboration between argenx and the teams of Profs. Pierre Coulie and Sophie Lucas of the Université catholique de Louvain (UCL), who “are recognized worldwide as leaders in the field of immuno-oncology,” with the company adding: “This collaboration has resulted in an innovative therapeutic approach to stimulate the immune system of cancer patients.”
The license agreement between argenx and Sopartec, the technology transfer office of the UCL, allowed argenx to sublicense its rights to the outcome of the collaboration with UCL. The terms of the sublicense agreement include upfront and milestone payments as well as royalties on future product sales.
The expertise and research of the teams of Coulie and Lucas, as well as their close collaboration with argenx, were said to be key for the conclusion of the agreement.
The collaboration between the teams of UCL and argenx started in 2013 in order to identify human antibodies that inhibit GARP function. A human anti-GARP antibody candidate has since been selected for further development.
In a joint statement, Coulie and Lucas commented that “This collaboration brings in the know-how of argenx and their partner AbbVie to carry our recently discovered approach of cancer immunotherapy into clinical development.”
“We are extremely pleased to collaborate with argenx and AbbVie in the clinical development and commercialization of immunotherapeutics,” said Philippe Durieux, CEO of Sopartec. “This collaboration is at the heart of our mission: detection and transfer of new technologies and innovative therapeutic interventions to combat diseases such as cancer.”
And here are some other bits of news from the world of oncology discovery, development and clinical trials:
ZIOPHARM shares favorable interim survival results for gene therapy candidate
BOSTON—ZIOPHARM Oncology Inc. on May 18 announced that interim results from the company’s ongoing Phase 1, multicenter dose-escalation study of the gene therapy candidate Ad-RTS-hIL-12 + orally administered veledimex in patients with recurrent or progressive glioblastoma would be presented at the American Society of Clinical Oncology Annual Meeting this June in Chicago. Ad-RTS-hIL-12 + veledimex is a novel viral gene therapy candidate for the controlled expression of interleukin 12 (IL-12), a pro-inflammatory cytokine critical for stimulating anticancer immune responses.
“Early results observed in the limited number of patients who have been treated with Ad-RTS-hIL-12 + veledimex are very encouraging for a Phase 1 study,” said Dr. Ennio Antonio Chiocca, Harvey W. Cushing Professor of Neurosurgery at Harvard Medical School, surgical director for the Center for Neuro-oncology at the Dana-Farber Cancer Institute and co-director of the Institute for the Neurosciences at Brigham And Women’s Hospital. “Virus-based gene therapy used to stimulate an immunological response in the brain is at the frontier of innovation in treatment, with Ad-RTS-hIL-12 offering perhaps the most controllable approach within this field. In this study, we see encouraging signs of immune activation following the administration of Ad-RTS-hIL-12 + veledimex.”
Overall Ad-RTS-hIL-12 + veledimex was well tolerated. All serious adverse events and grade 3 related toxicities were rapidly reversible upon discontinuation of veledimex. The most common related adverse events included headache, nausea/vomiting, fever, white blood cell/leukocyte count decrease, platelet count decrease and liver function test increase. Four subjects had related serious adverse events.
“Because the brain is a segregated and fragile environment, the ability to turn an immune response on and off is critical to balancing efficacy and tolerability,” said Dr. Francois Lebel, executive vice president of research and development and chief medical officer at ZIOPHARM. “As we continue to follow patients with extremely guarded prognoses in this multicenter trial, we hope that these promising early trends in survival are maintained. Our goal, once we reach an optimal dose, will be to promptly initiate registration trial discussions with regulators.”
“Basing our approach on the genetic engineering of adenovirus offers a simpler strategy than replicating viral-therapy options, one that can be rapidly controlled via a drug activating a gene switch, and one that does not rely on intratumoral catheterization,” added Dr. Laurence Cooper, CEO of ZIOPHARM. “We look forward to testing Ad-RTS-hIL-12 + veledimex not just on its own, but in combination with other immuno-oncology approaches, including checkpoint inhibitors and natural killer cells, in clinical trials that we plan to start this year and next. Our data suggest that patients can take a drug by mouth to activate the immune response in their tumors with exciting results.”
Horizon Discovery introduces high-throughput platform for immuno-oncology research
CAMBRIDGE, U.K.—Horizon Discovery Group plc announced in April the launch of a new platform for immuno-oncology research. The platform consists of several high-throughput assays that enable researchers to test thousands of potential anticancer drugs that recruit the body’s own immune system to fight cancer.
Horizon’s new platform is intended to enable researchers to study how new drugs are able to activate different types of immune cells as well as increase their ability to kill tumors. The platform includes high-throughput ADCC, CDC, primary T cell activation, mixed lymphocyte reaction and T cell-mediated tumor lysis assays in a 384-well format, allowing researchers to test hundreds of new drugs simultaneously.
“Horizon’s new immuno-oncology assays represent a great step forward in this important growth area, building on our greater-than 15 years of experience in high-throughput screening with over 10 million unique drug combinations screened,” said Dr. Edward Weinstein, president of services at Horizon Discovery. “The platform allows researchers to test thousands of new immuno-oncology drugs simultaneously, alone or in combination with other drugs. Immuno-oncology is an important focus area for Horizon, and we’re excited that these new assays will accelerate and streamline drug discovery, bringing highly effective new drugs to patients faster than before possible.”
This announcement follows the formation of Avvinity Therapeutics, a joint venture between Horizon and Centauri Therapeutics designed to capture the upside potential of Horizon’s intellectual property, platform technologies and capabilities in immuno-oncology in order to develop novel therapeutics. Horizon’s new immuno-oncology research platform will be deployed to support this effort as well as being offered commercially as a set of assay services to clients.
LOXO-101 shows durable antitumor activity across TRK fusion cancers
STAMFORD, Conn.—Loxo Oncology Inc., a biopharmaceutical company innovating the development of highly selective medicines for patients with genetically defined cancers, recently announced new results from its Phase 1 open-label, dose-escalation trial of LOXO-101, a selective inhibitor of tropomyosin receptor kinase (TRK). A presentation at the 2016 American Association for Cancer Research Annual Meeting in New Orleans in April provided updated data from the Phase 1 trial, which was last reported at the AACR-NCI-EORTC International Conference on Molecular Targets and Cancer Therapeutics in November 2015.
LOXO-101 Phase 1 study investigators reported that, as of the March 25, 2016, data cutoff date, 43 patients with solid tumors refractory to standard therapy had been enrolled and treated, including seven patients with cancers harboring TRK gene fusions. Data regarding the first three of these patients were initially reported at the AACR-NCI-EORTC conference in November 2015.
Six patients with TRK fusion cancers had been on the study sufficiently long for their first efficacy assessment, and all six exhibited significant tumor regressions. A seventh patient with a TRK fusion cancer was enrolled more recently and thus had not yet been evaluated for response as of the data cutoff date, though the patient remains on study. Five of the six efficacy evaluable patients achieved a confirmed partial response, as defined by standard RECIST criteria. The sixth patient demonstrated clear radiographic tumor regressions, including in the central nervous system, but has not met the threshold required for a RECIST response.
Tumor regressions have been observed in five different anatomically defined cancers: sarcoma, gastrointestinal stromal tumor, mammary analogue secretory cancer of the salivary glands, thyroid cancer and non-small cell lung cancer. No TRK fusion patients have progressed, with one patient in cycle 14, two patients in cycle 10 and three patients in cycle 7, as of the data cutoff date.
“The responses, durability and safety data with LOXO-101 clearly suggest that this is an important drug candidate for patients with TRK fusion cancers,” said Dr. David Hong, deputy chair and associate professor in the Department of Investigational Cancer Therapeutics at the University of Texas MD Anderson Cancer Center and presenter of the LOXO-101 oral presentation. “We are excited to continue to follow these Phase 1 patients and treat additional TRK fusion patients in the LOXO-101 Phase 2 trial. I hope these data encourage my colleagues to test for TRK fusions and refer patients to a LOXO-101 study.”
“The consistent efficacy of LOXO-101 in patients with TRK gene fusions, independent of tumor type, is very exciting,” added Dr. Josh Bilenker, CEO of Loxo Oncology. “The data update also provides an encouraging snapshot of response durability, particularly at the recommended Phase 2 dose of 100mg twice daily. These data suggest that LOXO-101 can deeply inhibit its target at a well-tolerated dose and generate durable disease control in a diverse group of patients with TRK gene fusions.”
Around the same time, Loxo also announced a confirmed RECIST partial response in the first patient enrolled in the recently opened pediatric Phase 1 dose-escalation trial of LOXO-101.
The peer-reviewed manuscript presenting this information described a 16-month old female patient with advanced infantile fibrosarcoma (IFS), a rare pediatric cancer. Genetic testing revealed an ETV6-NTRK3 fusion, which is frequently found in IFS. Following multiple unsuccessful surgeries and courses of chemotherapy, the patient was enrolled in the pediatric Phase 1 trial of LOXO-101, which employs a liquid formulation of the drug designed specifically for pediatric patients unable to swallow capsules. Her disease involved the neck, face, skull, mastoids and cervical vasculature. Throughout the first cycle of treatment with LOXO-101, the parents noted improved engagement and playfulness. At the end of cycle 1 (day 28), imaging of the brain and neck showed tumor regression of more than 90 percent from baseline. Repeat scans at the end of cycle 2 showed a continued decrease in tumor volume.
Bayer and Orion expand clinical development program in prostate cancer
WHIPPANY, N.J.—Recently, Bayer and Orion Corp. announced the expansion of the global clinical development program for the investigational androgen receptor (AR) antagonist BAY-1841788 (ODM-201) in the area of prostate cancer.
The Phase 3 study ARASENS will evaluate the compound in combination with standard androgen deprivation therapy (ADT) and the chemotherapy docetaxel in men with newly diagnosed metastatic hormone-sensitive prostate cancer who are starting first-line hormone therapy. The compound is currently in development for high-risk non-metastatic castration-resistant prostate cancer. The ARASENS clinical trial is expected to start enrolling patients towards the end of 2016.
“Although the treatment options for patients with advanced prostate cancer who are castration-resistant have evolved considerably in recent years, there were only few advances in the earlier, hormone-sensitive stage of metastatic disease. Continued research of investigational agents is needed to give patients more options,” said Dr. Joerg Moeller, a member of the executive committee of Bayer AG's Pharmaceutical Division and head of development. “Bayer’s commitment to prostate cancer research runs deep, and the initiation of ARASENS is the latest in a long line of studies designed to evaluate investigational compounds for prostate cancer.”
Bayer and Orion have recently expanded their 2014 agreement to include the joint development of BAY-1841788 (ODM-201) for mHSPC. Under that agreement, Bayer will commercialize the product globally; Orion has the option to co-promote BAY-1841788 in Europe, and is responsible for the manufacturing of the product.
“We believe that the profile of ODM-201 makes it an excellent candidate for further investigation in patients with metastatic hormone-sensitive prostate cancer. This is another example of how Orion's oncology research is committed to investigating compounds in prostate cancer. We are also happy to see Bayer's continuous commitment to ODM-201,” said Dr. Reijo Salonen, senior vice president of research and development at Orion.
ARASENS is planned to be initiated toward the end of 2016. Approximately 1,300 patients will be randomized (1:1 ratio) to receive either BAY-1841788 (ODM-201) or placebo in combination with an ADT of investigator's choice (LHRH agonist/antagonists or orchiectomy), started 12 weeks or less before randomization. Six cycles of docetaxel will be administered after randomization. The primary endpoint of the study is overall survival, and secondary endpoints include time to castration-resistant prostate cancer, time to initiation of subsequent antineoplastic therapy, symptomatic skeletal event-free survival, time to first symptomatic skeletal event, time to initiation of opioid use, time to pain progression, time to worsening of physical symptoms of disease and safety.
ARASENS adds to the ongoing clinical development program for BAY-1841788, which includes ARAMIS (NCT02200614), a randomized, Phase 3, multicenter, double-blind, placebo-controlled trial evaluating patients with non-metastatic castration-resistant prostate cancer who are at high risk for developing metastatic disease. ARAMIS is currently enrolling patients.
