MECHANISMS OF RESILIENCE
Understanding how malignant cells survive our attempts to kill them is critical for improving cancer therapy
BY DARRELL E. WARD
More than 1.6 million Americans were expected to develop cancer in 2012, and more than half of them were likely treated with radiation therapy. "Glioblastoma and certain other cancers are highly radiation resistant, while others such as neuroblastoma and certain lymphomas are inherently sensitive to radiation," says Timothy Lautenschlaeger, MD, an American Board of Radiology B. Leonard Holman Scholar in Radiation Oncology at The Ohio State University Comprehensive Cancer Center - Arthur G. James Cancer Hospital and Richard J. Solove Research Institute (OSUCCC - James).
"At this point, we don't really know why some types of cancer are so sensitive and others so resistant," Lautenschlaeger says. "It isn't simply a mutation or change in a particular gene that determines radiosensitivity in general; these differences are cancer-specific and dramatic."
The failure of radiation therapy - which uses high-energy X-rays to kill tumor cells - is a significant cause of the 577,000 deaths from cancer that were expected in 2012. "Overcoming the mechanisms of therapeutic resistance is key to improving cancer treatment," says Arnab Chakravarti, MD, chair of Radiation Oncology and co-director of the Brain Tumor Program at the OSUCCC - James.
Radiation resistance has multiple components that involve DNA repair, hypoxia, activation of prosurvival
pathways, cancer stem cells, and changes that lead to invasion and metastasis, he says. Chakravarti has recruited Lautenschlaeger and other leaders in the field to the OSUCCC - James in an effort to understand and overcome radiation resistance.
FEN XIA, MD, PhD
associate professor of Radiation Oncology and an OSUCCC - James researcher
who specializes in DNA damage response and repair in radiation resistance.
Radiation kills cancer cells mainly by causing breaks in both strands of the DNA helix. "Failure to repair double-strand breaks triggers cell-death processes, including apoptosis, whereas tumors adept at repairing double-strand breaks tend to be radiation resistant," says Fen Xia, MD, PhD, associate professor of Radiation Oncology and an OSUCCC - James researcher who specializes in DNA damage response and repair in radiation resistance.
Xia wants to understand how cells sense DNA damage and initiate the repair response, and how these actions differ between tumor and normal cells. "We want to interrupt this process and kill tumor cells while preserving normal cells," she says.
To overcome radiation resistance, Xia is exploring two strategies: protecting normal cells from the lethal effects of radiation while concurrently enhancing radiationinduced tumor cell death.
A 2009 study by Xia published in the Journal
of Clinical Investigation is a step toward the first strategy. The laboratory
and animal study suggests that lithium can enhance a key DNA repair process
called non-homologous end-joining repair in irradiated noncancerous neuronal
BRCA1 AND PARP1 INHIBITION
An example of the second strategy is Xia's work to make cancer cells with functional BRCA1 protein susceptible to PARP1 inhibition. PARP1 - or poly ADPribose polymerase - is a molecular complex that quickly repairs breaks in one of DNA's two strands.
DNA single-strand breaks can lead to double-strand breaks during replication. Unrepaired double-strand breaks are lethal to proliferating cancer cells. Dysfunction in the repair of both single-strand breaks and doublestrand
breaks will be synthetically lethal.
BRCA1 - infamous, along with BRCA2, for raising breast-cancer risk when mutated - is a DNArepair protein that corrects doublestrand breaks. Xia, in a 2012 study published in the journal Cancer Research, and others have shown that the PARP1 inhibitor olaparib is highly selective in killing BRCA1-mutated familial breast tumors.
"Unfortunately, over 90 percent of patients who develop sporadic breast cancer carry functional BRCA1 and BRCA2 proteins and are proficient in repair of doublestrand breaks, precluding them from this potent therapy," Xia says.
"An important goal of our work is to make PARP inhibitors available to the majority of patients, those with working BRCA1 genes."
In earlier work, Xia demonstrated that BRCA1's repair function occurs in the nucleus, but that the BRCA1 protein shuttles between the nucleus and cytoplasm. In the cytoplasm, it loses its DNA-repair function and instead facilitates apoptosis. "We show that if the BRCA1 protein moves to the cytoplasm, cells die much more quickly and efficiently," Xia says.
Xia and her collaborators believe that if they can force BRCA1 out of the nucleus, it could make tumors that have functional DNA repair susceptible to PARP-inhibition therapy. The same strategy could be used to sensitize tumors to radiation-induced DNA damage, she says.
RESEARCH BY NICHOLAS DENKO, MD, PHD,
is supported by NCI program project grant CA067166-15 and NCI grants CA156950-02 and CA163581-01A1.
When tumors outgrow their blood supply, they develop new blood vessels that are abnormal and leaky. These poorly perfused tumors have pockets of low oxygen and necrosis; that is, they are hypoxic.
Hypoxia within tumors reduces the effectiveness of radiation therapy. When therapeutic ionizing radiation strikes cellular molecules, it generates ions, free electrons and hydroxyl radicals. Some of these reactive elements are captured by the cell's natural buffering system, but many strike its DNA and cause double-strand breaks that overwhelm the repair system and kill the cell.
"The presence of oxygen enhances the fixation of this DNA damage and the lack of oxygen inhibits it," says Nicholas Denko, MD, PhD, associate professor of Radiation Oncology at the OSUCCC - James and an authority on hypoxia and radiation therapy.
The absence of oxygen in tissue can decrease the effective radiation dose by a factor of three - defined as the oxygen enhancement ratio. That is, it can take three times the radiation dose to kill seriously hypoxic cells versus well oxygenated cells, he says.
But there is little latitude for raising a patient's radiation dose, which is chosen primarily to kill tumor cells and is about the maximum possible without causing complications.
"Most tumor types have areas of hypoxia," Denko says. Studies have been done in patients with head and neck cancer, cervical cancer or softtissue sarcomas. "We calculate that in these tumors even a small fraction of hypoxic tumor cells can have a big effect on patient outcome," he says.
"People have understood the oxygen enhancement ratio for 50 years, but no one has figured out how to use it clinically," he explains. Investigators have tried to deliver more oxygen to the tumor. They've transfused patients with red blood cells, given patients erythropoietin to grow more red blood cells and had patients breathe pure oxygen to get more oxygen to the tumor.
"These have all had disappointing clinical effectiveness," he says.
Denko and his colleagues are approaching the problem differently. "Rather than trying to increase the tumor's oxygen supply, we want to reduce the tumor's oxygen demand," he says.
The researchers are evaluating drugs that reduce oxygen consumption by mitochondria, the main oxygen sink in cells. "Mitochondria produce ATP, which powers everything we do," Denko says. "There are drugs that inhibit mitochondria as a side effect. If we can use them in tumors to reduce oxygen demand by mitochondria, we might improve the tumor's response to radiation therapy."
“We have a lot more work to do before attempting this in
patients, but the idea is to give the drug first, allow the tumor oxygen
concentration to rise, then irradiate the tumor,” he says. “If the strategy increases
treatment effectiveness in tumor models, it could be moved quickly into
clinical use because the drugs we are studying have already been approved by
the Food and Drug Administration (FDA).”
Research by Ohio State's Arnab Chakravarti shows that radiation therapy can activate critical pro-survival signal-transduction pathways and shut down cell-death pathways, enhancing cancer-cell survival and radiation resistance.
In a Journal of Clinical Oncology study, Chakravarti reported that activation of the PI3 kinase/AKT pathway is associated with adverse outcomes in malignant glioma patients. These pro-survival molecules were activated far more often in higher-grade gliomas. "Glioblastomas had the highest activation of PI3 kinase family members compared with grade III tumors or grade II tumors, and the degree of activation was strongly associated with adverse clinical outcome and radiation resistance," Chakravarti says. As such, the degree of pathway activation in glioblastoma is an independent prognostic marker over and beyond tumor grade.
Loss of the PTEN
tumor-suppressor gene also constitutively activates AKT signaling and enhances
survival in gliomas and other tumor types, he notes.
In a 2011 New England Journal of Medicine paper, Chakravarti collaborated with Marcus Bredel, MD, an associate professor at the University of Alabama Birmingham and an adjunct associate professor of Radiation Oncology at the OSUCCC - James, and showed that loss of a gene might improve survival for certain glioblastoma patients.
"The NF-kB pathway is thought to be related to the PI3 kinase/AKT pathway, so it's interesting that an NF-kBIA deletion is also strongly associated with therapeutic resistance," Chakravarti says.
"Overall, this shows that these prosurvival pathways play a major role in mediating radiation resistance.
ARNAB CHAKRAVARTI, MD
is supported by: NCI grants CA108633 and CA148190, and by the Brain Tumor Funders Collaborative.
Erica Hlavin Bell, PhD, research assistant professor of Radiation Oncology at the OSUCCC - James, works closely with Chakravarti to identify biomarkers of radiation resistance in low-grade gliomas and prostate cancer.
"We want to learn whether grade 2 and grade 3 gliomas respond differently to treatment compared with grade 4 tumors (i.e., glioblastoma multiforme)," Bell says.
"Most studies of brain tumors have been completed with varying grades, and patients with different grades are often treated in a very similar fashion," she adds. "We are asking whether lower grade tumors need the same treatment as glioblastoma. We've learned that, at the molecular level, lower-grade brain tumors are very different from grade 4 tumors. Because of this, we believe that their treatment response is probably also different. We want to understand that in detail."
Bell's investigations use tumor samples obtained through two Radiation Therapy Oncology Group (RTOG) trials: RTOG- 9813 (ClinicalTrials.gov identifier NCT00004259) and RTOG-9802 (ClinicalTrials.gov identifier NCT00003375). She is looking at low-grade glioma tissue for new and known biomarkers and comparing them with patient outcomes.
Specific biomarkers of
known importance for lower-grade gliomas include 1p19q co-deletion – patients
with that deletion tend to do better, Bell says – and patients with mutated IDH1
also tend to do better. MGMT methylation status is also noted. (MGMT is a
DNA-repair gene; when this gene is silenced by methylation in a tumor, patients
tend to have better outcomes.)
"This tissue is precious," she says, "and we want to acquire as much information as possible from each specimen."
“These are rare tumors, accounting for less than 1 percent of
all malignancies, and it would be challenging to perform studies of low grade
gliomas at The Ohio State University alone,” Bell says. “We really need
national and international clinical trials to get enough tissue to produce a
statistical result that allow you to make a clinical decision.”
Chakravarti and Bell are also conducting a study in
collaboration with the University of Freiburg, Germany, to identify novel
biomarkers of prostate cancer utilizing prostatectomy specimens of patients who
received radiation therapy after prostatectomy. The goal of this study is to
identify markers of radiation response in prostate cancer.
Bell’s novel studies include global microRNA profiling using
the Nanostring platform. They also use a specific Affymetrix microarray called
OncoScan to perform a SNP/CNV analysis that is designed to work with formalin-fixed,
paraffin-embedded (FFPE) tissue.
“We’re also working with our sequencing core to establish a sound
protocol for Methylseq, whole genome methylation sequencing using FFPE tissue.
In addition, Bell has designed a translational study intended to improve
the treatment of non-small cell lung cancer that has a mutated SMARCA4 gene. The
mutation affects chromatin remodeling complexes that are important for DNA
repair, Bell says. She’s investigating whether the mutation alters DNA repair
capacity and how that might be exploited therapeutically. “If the mutation
decreases DNA repair, those patients might respond better to radiation,” she
says. This work is supported by the Lung Cancer Research Foundation.
GLIOBLASTOMA AND MELANOMA
People with grade-4 melanoma of the skin or with grade-4 glioma (glioblastoma) in the brain survive about one year after diagnosis. Both malignancies are highly resistant to chemotherapy and radiation. "We are trying to identify the cause of that resistance," says Kamalakannan Palanichamy, PhD, research assistant professor of Radiation Oncology at the OSUCCC - James.
In both malignancies, Palanichamy and his lab are combining genomic, transcriptomic, epigenomic, proteomic and metabolomic analyses to identify subsets of patients who will better benefit from a particular treatment.
glioblastoma cell lines, Palanichamy and his colleagues blocked the AKT pathway
and identified a set of cell lines that became more radiation sensitive. In
another subset of cell lines, the combination did not increase the number of
“Understanding the molecular differences between these two
groups could help us improve treatment by identifying glioblastoma patients who
are likely to better respond to radiation therapy,” he says.
Using glioblastoma tissue from
patients, Palanichamy and his colleagues have identified a gene called MGMT that
encodes a DNA-repair protein and that when inactive indicates greater
sensitivity to radiation. “In 40 percent of cases the gene is methylated, so
these patients have improved survival outcomes and may derive greatest benefit from standard radiation and chemotherapy,”
he says. “In the remaining 60 percent of patients, the gene is functional,
damaged DNA is repaired and radiation has more marginal benefit.”
Patients with the intact gene receive a different combination
therapy, or an inhibitor that is not associated with DNA repair.
his group is devising strategies to target BRAF-mutant melanomas. The drug vemurafenib
(PLX4032) is given to this subset of melanoma patients. Patients initially
respond well to the drug, but then develop resistance to it through up-regulation
of cell survival mechanisms. In an effort to overcome the problem, Palanichamy’s
group is evaluating sequential monotherapies using PLX4032 sensitive and
In addition, Palanichamy is investigating the role of self-renewing populations of cancer cells, also called cancerstem cells (CSCs), in radiation resistance. They isolate CSCs from glioblastoma tumors, culture them and expose them to chemotherapy, small-molecule inhibitors and radiation alone and in combination.
"We have identified a set of novel antigens that pass signals through the STAT pathway
and are associated with resistance, and we are investigating ways to inhibit
them,” Palanichamy says.
RECURRENT BLADDER CANCER
Some 73,500 Americans were expected to develop bladder cancer in 2012, and an estimated 14,900 people died from the disease. Three quarters of newly diagnosed bladder-cancer cases present with superficial, noninvasive tumors. These are treated by minimally invasive surgery through the urethra, sometimes with chemotherapy. There is a high rate of prolonged survival.
For invasive bladder cancer, the bladder is frequently removed. In many cases, however, it is possible to preserve the bladder with transurethral surgery, radiation and chemotherapy, says Lautenschlaeger, who is studying radiation therapy for bladder cancer.
"Bladder removal versus bladder preservation is also important when superficial bladder cancer recurs as muscle-invasive cancer," Lautenschlaeger says.
Lautenschlaeger is working to identify biomarkers of recurrence that will enable urologists to identify patients who are better candidates for surgery or for bladder-preservation therapy in collaboration with a team led by William U. Shipley, MD, at the Massachusetts General Hospital (MGH)/Harvard Medical School. Together with MGH investigators who originally pioneered the concept of bladder-preservation therapy, Lautenschlaeger leads a correlative study for markers of treatment resistance that is part of a phase II randomized study for patients with muscle-invasive bladder cancer evaluating transurethral surgery and concomitant chemoradiation (Clinical trials.gov identifier NCT00777491).
The trial compares
chemotherapy and radiation therapy regimens in patients with muscle-invasive
bladder cancer removed by transurethral surgery. Lautenschlaeger’s study compares
molecular profiles of patients who are disease free after treatment with those from
patients with recurrent cancer to identify potential biomarkers of resistance.
"Ideally, we will one day be able to tell patients whether or not they are good candidates for bladder preservation," he says. "Patients with markers indicating bladder preservation can avoid the quality-of-life changes associated with bladder loss, and patients with markers indicating a high risk of recurrence can avoid the side effects of chemotherapy and radiation."
Deliang Guo, PhD, assistant professor of Radiation Oncology, is investigating links between tumor-cell metabolism cells and radiation and chemotherapy resistance.
"Cancer cells have altered and enhanced metabolism, and oncogenes are involved in that metabolic reprogramming," Guo says. "We want to unravel the links between oncogenic signaling pathways and cancer metabolism."
Additionally, Guo hypothesizes that radiation therapy can alter glucose metabolism in cancer cells in ways that contribute to radiation resistance. He reasons that radiation can boost the production of ATP and of nucleic acids, lipids and amino acids, which cancer cells need for growth and proliferation.
He notes that laboratory studies have shown that radiation
can increase activity of the oncogenic PI3 kinase/AKT pathway. “This pathway
can maintain tumor growth and survival,” he says.
Still other studies show that increases in PI3 kinase/AKT signaling can increase
glucose uptake, boost ATP generation, and increase the synthesis of lipids, DNA
and amino acids. “Our lab has shown that activation of the PI3 kinase/AKT
pathway can increase glucose uptake and ATP generation, and that high glucose
uptake can contribute to the synthesis of lipids, DNA and amino acids,” he
"We are now investigating whether the irradiation of cancer cells directly increases glucose uptake and ATP production in cancer cells," he says.
Guo, whose specialty is oncogene signaling and metabolic pathways, is collaborating on this work with Chakravarti, a specialist in radiation therapy. "We've combined our expertise to learn whether radiation causes this metabolic change, and whether we can interrupt this interaction and reduce radiation resistance," Guo says.
Research is improving radiation therapy, Lautenschlaeger says. "We're learning, for example, that changing the fractionation schedule might improve treatment outcomes in certain tumors. The traditional schedule of low daily doses works well and is safe, but some cancers might be more sensitive to one or just a few high doses of radiation.
"There are now protocols for lung cancer that use only a few fractions of very high doses of radiation," he says. "This was first tried in patients with advanced, inoperable disease. Now, trials using radiation without surgery are available to patients who are operable.
"The day is coming when we will do molecular profiling of tumors and understand that the tumor has a particular array of gene changes, and that information will enable us to determine the optimal treatment regimen," he says. "Radiation therapy for cancer will become much more personalized and even more effective."
Read this story online for more details about the research described here.