A sluggish march toward cure for pancreatic cancer

Early detection may not be helpful because treatment has been largely unsuccessful.

Physicians have made great strides in treating cancer and extending patient survival, but that same success is not evident in pancreatic cancer, the fourth-leading cause of cancer deaths in the US.

Five-year OS has gone from 3% from 1975 to 1977 to only 6% from 1995 to 2000, according to “Cancer Statistics, 2010,” published last year in CA: A Cancer Journal for Clinicians. Median OS is less than 6 months.

Tanios Bekaii-Saab, MD, medical director of gastrointestinal oncology at The Ohio State University Comprehensive Cancer Center, said the RAS gene is a major target for treating pancreatic cancer. Photo Courtesy of Ohio State University

“There is not much to talk about in terms of treatment these days, unfortunately,” said Tanios Bekaii-Saab, MD, medical director of gastrointestinal oncology at The Ohio State University Comprehensive Cancer Center — Arthur G. James Cancer Hospital and Richard J. Solove Research Institute. “More work has to be done in pancreatic cancer. There is less funding for pancreatic cancer research than any other cancer of significance. We need more research into understanding the biology, the genetics, more research into getting patients into early detection studies and more research to find better therapeutics and agents that will actually make a difference across stages.”

Only surgery has been shown to have any curative effect. Treatments proven effective in other cancers, such as bevacizumab (Avastin, Genentech) and cetuximab (Erbitux, Imclone), have failed in pancreatic cancer, so researchers continue working on new drugs and pathways of treatment to attack the disease. In general, however, treatment has advanced little in this disease, leading to an increased interest in early detection programs and prognosis markers.

Screening

Approximately 80% to 85% of patients present with advanced pancreatic cancer. Bekaii-Saab said only 10% of the remaining patients have curable disease. It stands to reason, then, that earlier diagnosis of the disease could improve OS; several presentations made at the 2011 Gastrointestinal Cancers Symposium addressed this topic.

In results presented by Nicolai A. Schultz, MD, with the department of surgical gastroenterology at Rigshospitalet in Denmark, researchers found that 83 microRNAs were differentially expressed in pancreatic cancer compared with normal tissue and 32 microRNAs were differentially expressed in cancer compared with chronic pancreatitis. The most differentially expressed microRNAs were miR-614, miR-492, miR-622, miR-135b and miR-196.

Schultz said results validate the findings of earlier studies, although they still require further validation. However, microRNAs could represent a biomarker to detect early-stage pancreatic cancer.

David K. Chang, MBBS, with the New South Wales Pancreatic Cancer Network, said researchers derived a nomogram using clinicopathologic variables and aberrant expression of S100A4 and S100A2. After evaluating the association between aberrant S100A4 calcium-binding protein expression and survival in 372 patients who underwent surgical resection, researchers found that high S100A4 expression was an independent factor for poor prognosis.

Chang said S100A4 was an independent prognostic factor for both the training set (HR=5.00; 95% CI, 2.29-10.9) and the validation set (HR=1.78; 95% CI, 1.29-2.46).

Researchers stratified the cohort into three prognostic groups based on aberrant expression of S100A4 and S100A2. Chang said a preoperative nomogram using only variables that could be measured preoperatively, such as tumor size and molecular biomarkers, predicted survival more accurately than nomograms derived from using clinicopathologic variables.

“This may be used to improve the accuracy of prognostic nomograms and to help us better prognosticate and predict prognosis preoperatively,” he said. “This will hopefully help us select patients for more appropriate and personalized therapy so that we only operate on the right patients, and as a consequence, improve overall outcomes.”

There is an interest in screening for pancreatic cancer and developing new risk predictors because, so far, treatment has not extended survival appreciably, Bekaii-Saab said. The potential problem with focusing on detection is that uncovering early pancreatic cancer and determining who is likely to progress may not be helpful because treatment has been largely unsuccessful, said William Cance, MD, chair of surgical oncology at Roswell Park Cancer Institute in Buffalo, N.Y., and chief scientific officer for CureFAKtor Pharmaceuticals.

“Are we going to remove an entire pancreas just because of potential risk of developing cancer? That’s not an operation many surgeons would contemplate,” Cance said. “We have to address the therapeutics. The overall survival is 2% to 5%; we don’t need to subset those patients. We need the therapeutics. If you give me a 30-year-old patient with a 70% chance of developing pancreatic cancer during their lifetime, taking out the whole pancreas now would severely impact that patient’s quality of life and morbidity from non-cancer causes related to the pancreatectomy. There’s a real risk with the screening and the prognostication.”

H. Shep Wild, president and CEO of CureFAKtor, also attended the symposium and said he was “startled” to see so much attention going toward the question, “How much longer is this patient going to live?”

“Except in extraordinary cases, the answer is ‘not long,’ tragically,” Wild said. “What’s needed here is the therapeutics.”

Controversy over VEGF

CureFAKtor manufactures CFAK-C4 and a new compound called CFAK-Y15. The FDA granted orphan drug status to CFAK-C4 in January. The company is planning a phase 1 study for 2012 evaluating the drug combined with gemcitabine for patients with pancreatic cancer, with preliminary results expected within 6 months to 12 months. CFAK-Y15 is currently in preclinical testing.

William Cance

Cance said focal adhesion kinase (FAK) is a binding site for several types of oncogenic proteins, such as HER-2 and P53, and angiogenesis receptors, such as VEGF-3. He said FAK also acts as a kind of “force field” that protects tumor cells until they metastasize and attack the patient.

To disrupt that binding and increase chemosensitivity, researchers working at the University of Florida discovered CFAK-C4, a small-molecule allosteric FAK inhibitor.

“CFAK-C4 sits in the pocket where VEGF-3 binds to FAK and activates it,” he said. “That allows chemotherapy, in this case gemcitabine, to be much more effective against pancreatic cancer.”

In study results presented at the 2011 Gastrointestinal Cancers Symposium, VEGF-3 inhibition with 10 mcm of CFAK-C4 led to an 80% reduction in the viability of pancreatic cancer cells, a 50% reduction in motility and a decrease in proliferation.

Additionally, CFAK-C4 was associated with apoptosis in cancer cells and increased the chemosensitivity of pancreatic cancer cells in vitro.

“One of the unique things about C4 is that it’s an antihistamine; it’s been in use in Eastern Europe and Russia,” Cance said. “Nobody realized it was an anticancer drug until we came along, and through our structural-based analyses determined that it was a potent inhibitor of these kinases. It’s a known drug, and that’s a good thing because we know it has a safe toxicity profile.”

However, phase 3 results published Feb. 8 online in Lancet Oncology, by Kindler and colleagues, concluded that “targeting of VEGF signaling is an ineffective strategy in this disease.”

In a double blind, placebo-controlled study funded by Pfizer, 305 patients were randomly assigned to gemcitabine plus axitinib, a selective oral inhibitor of VEGF receptors 1-3. Another 308 patients were assigned to gemcitabine plus placebo.

All patients were diagnosed with advanced pancreatic adenocarcinoma and treated from July 2007 to October 2008. About three-quarters of patients in both groups had metastatic disease, and roughly 50% of patients had an ECOG performance status of 1.

At a planned interim analysis in January 2009, the Independent Data Monitoring Committee determined the futility boundary had been crossed and stopped the study. Treatment assignments were unmasked, and researchers recommended discontinuation of axitinib.

Median follow-up was 27 weeks in the gemcitabine plus placebo group and 27.4 weeks in the gemcitabine group. Median OS was 8.5 months (95% CI, 6.9-9.5) in the experimental arm and 8.3 months (95% CI, 6.9-10.3) in the gemcitabine and placebo arm (HR=1.014; 95% CI, 0.786-1.309). Median PFS of 4.4 months was the same for both treatment groups (HR=1.006; 95% CI, 0.779-1.298).

“This randomized phase 3 trial clearly shows that the addition of axitinib to gemcitabine does not improve survival in patients with locally advanced or metastatic pancreatic cancer,” the researchers concluded. “These data also confirm the findings obtained in previous phase 3 studies of the VEGF inhibitors bevacizumab and aflibercept that inhibition of this pathway is ineffective in patients with this disease. … These results add to increasing evidence that targeting of VEGF signaling is an ineffective strategy.”

Cance said there is a role for targeting angiogenesis in pancreatic cancer, and he will continue research into clinical application for its compounds.

“It’s well documented that angiogenesis plays an important role in the progression of cancer. There are different ways to target angiogenesis. This [Kindler] study has used kinase enzyme inhibitors that span the vascular epithelial growth factor receptors as a way to target the angiogenesis. We’ve targeted FAK, which has shown to be a target to mediate angiogenesis,” he said. “We target not the kinase but the interactome. We make these alloseteric inhibitors that target the interaction between the angiogenic receptor and FAK and we are seeing excellent results in preclinical models. We certainly will keep developing this.”

Extended survival with FOLFIRINOX

FOLFIRINOX, a combination of 5-FU plus leucovorin, irinotecan and oxaliplatin, is drawing interest because it was associated with extended PFS and OS, according to interim results from the phase 3 PRODIGE 4/ACCORD 11 trial. Thierry Conroy, MD, with the department of medical oncology at Centre Alexis Vautrin in Vandœuvre-lès-Nancy, France, presented results from a planned interim analysis at the 2010 ASCO Annual Meeting.

In that trial, which was stopped early by the Independent Data and Safety Monitoring Committee, 342 patients with metastatic pancreatic adenocarcinoma were assigned to gemcitabine (n=171) while the rest were assigned to FOLFIRINOX.

Median PFS was 6.4 months in the FOLFIRINOX group vs. 3.4 months for gemcitabine (HR=0.47; 95% CI, 037-0.59). At 26.6 months’ follow-up, median OS also favored the combination, 11.1 months vs. 6.8 months.

The confirmed response rate for the experimental arm was 27.6% compared with 10.9% for gemcitabine (P=.0008). Disease control rate was 70.2% for FOLFIRINOX compared with 50.9% for gemcitabine. Similarly, duration of control favored the FOLFIRINOX group, 5.9 months vs. 4 months.

FOLFIRINOX was also associated with increased rates of grade-3/grade-4 toxicities for diarrhea (12.3% vs. 1.6%); nausea (15.6% vs. 6.3%); vomiting (17.2% vs. 6.3%); fatigue (24% vs. 14.3%); neutropenia (47.9% vs. 19.2%) and febrile neutropenia (5.7% vs. 0%).

Conroy said FOLFIRFOX was associated with a 53% reduction in risk for progression and delayed quality-of-life degradation.

“This is the first time that, in a phase 3 randomized study, we have achieved 11 months median survival for metastatic pancreatic cancer,” Conroy said in his presentation. “We recommend FOLFIRFOX as the new international standard of care [for] patients with metastatic pancreatic cancer, but only in patients with normal levels of bilirubin or performance status of zero or one.”

“The triple regimen showed a little bit less than a doubling of survival that unfortunately comes at the price of significant toxicities,” Bekaii-Saab said. “That tells you there is hope for patients and drug development in pancreas cancer, if you can find the right agent(s) and the right target.”

Targeting the RAS pathway

Bekaii-Saab’s research is part of a two-armed, randomized phase 2 study of carboplatin and paclitaxel plus Reolysin vs. carboplatin and paclitaxel alone as first-line treatment for patients with recurrent or metastatic pancreatic cancer. Reolysin is a proprietary formulation of the human reovirus, a harmless, naturally occurring virus usually found in the throat and airway passages, developed by Oncolytics Biotech in Calgary.

“As it gets into cells that carry a mutation in the RAS gene, it becomes activated and lytic to those cells that carry the mutation,” Bekaii-Saab said. “RAS remains one of the main targets for treating pancreatic cancer. Everyone recognizes the importance of this target. The problem is finding agents, or a virus in this case, that can actually reach the target and kill the cancer cell with the activated pathway.”

He said as with any foreign agent, the immune system will move to attack Reolysin once it is injected. Accordingly, researchers are delivering Reolysin combined with carboplatin and paclitaxel.

Bekaii-Saab said chemotherapy combination works well with Reolysin because it suppresses the component of the immune system that clears the virus.

Reolysin works by invading a tumor cell with an activated RAS pathway and replicating itself until it bursts the walls of the tumor cell. Bekaii-Saab said the RAS mutation appears in up to 90% of pancreatic cancer cells, and Reolysin appears to destroy tumor cells without affecting healthy cells. Because reovirus is present in most people without causing any symptoms, patients should not experience additional adverse effects.

“The virus itself, which carries no major toxicities that we know of, is innocuous to the body,” he said. “But the virus has to reach the cancer cell, so we have to give chemotherapy. We can’t escape toxicities from the chemotherapy, but those toxicities probably won’t be amplified by the virus.”

Accrual has started, with a goal of eventually enrolling 70 patients at OSU Comprehensive Cancer Center, Georgetown Lombardi Comprehensive Cancer Center and Memorial-Sloan Kettering Cancer Center. Full accrual is expected within 18 months, and Bekaii-Saab said he hopes to have preliminary results within 2 years.

Reolysin is also under investigation in head and neck, ovarian and lung cancers, and Bekaii-Saab said OSU will be taking part in a study evaluating the drug’s efficacy in colon cancer.

“This would be the first oncolytic virus that is showing significant promise,” he said. “There have been a number of oncolytic viruses developed over the years. None of them made it anywhere. This one seems to be the one virus to show some interesting promise that could one day be applied in the clinic.”

Diane M. Simeone

Attacking the disease at its origin

Researchers at the University of Michigan Comprehensive Cancer Center were the first to discover pancreatic cancer stems cells in 2007. Now, the goal is to turn that discovery into something of clinical value, said Diane M. Simeone, MD, chief of gastrointestinal surgery.

“We’re continuing to try to understand the basic biology of pancreatic stem cells in the lab,” she said. “We’re also doing preclinical studies in the lab to figure out an optimal way of targeting those cells.”

Simeone said researchers at the University of Michigan are part of a study exploring the use of Genentech’s GDC-0449 alone or combined with gemcitabine to target the hedgehog signaling in patients with metastatic disease. The school is also participating in another trial using a gamma secretase inhibitor from Roche to attack the notch signaling pathway in patients with resectable pancreatic cancer.

Unlike breast cancer, in which the existence of progenitor stem cells is well defined, Simeone said the cell of origin is still unclear in pancreatic cancer, but the stem cells of this disease are highly resistant to standard therapy.

“To eradicate the disease, we’re going to have to come up with therapies to target stem cells,” she said. “We think those therapies will have to be used in combination with other therapies that target the bulk of the cancer cells. It might be that we find something effective at killing both, but our preclinical work suggests that we’ll need to use combination therapy to get the best results.”

Simeone said promising results from phase 2 trials often turn disappointing in phase 3 outcomes. The key for researchers is to better understand the biology of the disease.

“Pancreatic cancer is not a single disease but, probably, there are subtypes, and these subtypes are more subtle than what we see with breast cancer, for example. Some of these treatments that demonstrate a response are probably due to selection bias in the results, but certainly, for a number of these therapies, often there is a subset of patients who do have a response,” she said. “Moving forward … we need to better understand if there are subtypes of pancreatic cancer or genetic predictors to responsiveness.” – by Jason Harris

Disclosures: Dr. Cance is a shareholder in CureFAKTor Pharmaceuticals, LLC and has a direct financial interest in CureFAKtor Pharmaceuticals products CFAK-C4 and CFAK-Y15.

For more information:

• Bao Q. Cancers. 2010;2:1629-1641.
• Bilimoria KY. Cancer. 2007;110:1227-1234.
• Chang DK. #154. Presented at: the 2011 Gastrointestinal Cancers Symposium; Jan. 20-22, 2011; San Francisco.
• Conroy R. #4010. Presented at: the 2010 ASCO Annual Meeting; June 4-8, 2010; Chicago.
• Cunningham D. J Clin Oncol. 2007;27:5513-5518.
• Heinemann V. J Clin Oncol. 2006;24:3946-3952.
• Hermann PC. Cell Stem Cell. 2007;1:313-323.
• Hochster HS. Cancer. 2006;107:676-685.
• Jemal A. CA Cancer J Clin. 2010;60:277-300.
• Kindler HL. J Clin Oncol. 2010;28:3617-3622.
• Kindler HL. Lancet Oncol. 2011;doi:10.1016/S1470-2045(11)70004-3.
• Kurekova EV. #214. Presented at: the 2011 Gastrointestinal Cancers Symposium; Jan. 20-22, 2011; San Francisco.
• Lee CJ. J Clin Oncol. 2008;26:2806-2812.
• Li C. Cancer Res. 2007;67:1030-1037.
• Philip PA. J Clin Oncol. 2010;28:3605-3610.
• Philip PA. Lancet. 2011;doi:10.1016/S1740-2045(11)70029-8.
• Schultz NA. #153. Presented at: the 2011 Gastrointestinal Cancers Symposium; Jan. 20-22, 2011; San Francisco.
• Wolpin BM. J Natl Cancer Inst. 2009;101:424-431.

Is VEGF a valid treatment target in pancreatic cancer?

.

The randomized phase 3 trial of gemcitabine plus axitinib is now the fourth largest randomized phase 3 trial evaluating the addition of an anti-vascular agent to a gemcitabine or gemcitabine-based backbone, which is negative.

Eileen M. O’Reilly

Two studies have evaluated bevacizumab (CALGB 80303, AViTA); a third, aflibercept; and the current study, the broad spectrum VEGF inhibitor axitinib. The collective evidence from the clinical experience totaling more than 2,400 patients, despite compelling preclinical data and rationale, is that inhibition of VEGF in advanced pancreas cancer is a failure.

Why this is so in pancreas adenocarcinoma in contrast to other solid tumor malignancies is not clearly explainable. This study also highlights another area where clinical research in pancreas adenocarcinoma has struggled, and that is what is an acceptable signal for moving from a phase 2 to a phase 3 trial? The current phase 3 study was designed on the basis of a relatively modest signal from a randomized phase 2 trial. Whether randomized phase 2 designs or simply larger phase 2 designs are used, it is clear that the signal to proceed to a phase 3 needs to be stringent.

Eileen M. O’Reilly, MD, is an associate member of the Memorial Sloan-Kettering Cancer Center and associate professor of the Weill Medical College of Cornell University.

Disclosure: Dr. O’Reilly reports no relevant financial disclosures.

Because VEGF and its receptors are thought to play a key role in tumor angiogenesis, most of the antiangiogenic agents currently being tested target the VEGF ligand-receptor system.

Almost all preclinical studies targeting this system showed significant reduction of tumor size and reduced angiogenesis. However, most human phase 3 trials using antiangiogenic therapies, including this [Kindler] study, have produced poor or disappointing results.

Multiple factors could contribute to failure of the antiangiogenic therapy in pancreatic cancer.

Hwyda A. Arafat

Although some studies have reported a positive correlation between VEGF and VEGF receptor levels, blood vessel density and disease progression, others did not. This might be due to the fact that pancreatic cancer is not a grossly vascular tumor.

In addition, pancreatic cancer overexpresses multiple additional mitogenic growth factors, which are also angiogenic, such as epidermal growth factor, transforming growth factor-alpha, hepatocyte growth factor and fibroblast growth factors. Therefore, although VEGF appears to be of crucial importance in promoting the growth and metastasis of pancreatic cancer cells, other factors are most likely also involved in this process.

Another interesting aspect is that tumors subjected to sustained VEGF signaling blockade may partly adapt by increasing the expression of vascular basement membrane collagen IV, heparan sulfate proteoglycan and heparanases, resulting in the sequestration and release of VEGF in the immediate microenvironment of the tumor vessel.

Consequently, VEGF receptor activation persists and endothelial survival signaling is stimulated. So even tumors that are initially highly responsive to VEGF blockade could rescue VEGF signaling by increasing the microenvironmental bioavailability of VEGF.

Thus, strategies that target this response and the other mitogenic-angiogenic factors may enhance the efficacy of anti-VEGF therapy in patients with pancreatic cancer.

Hwyda A. Arafat, MD, PhD, is an associate professor in the departments of surgery and pathology anatomy and cell biology at Jefferson Medical College, Kimmel Cancer Center, Philadelphia.

Disclosure: Dr. Arafat reports no relevant financial disclosures.