In this podcast, Dr. Jeffrey Meyerhardt discusses some of the research on gastrointestinal, or GI cancers presented at ASCO’s 2015 Annual Meeting.
ASCO: You're listening to a podcast from Cancer.Net. This cancer information website is produced by the American Society of Clinical Oncology, known as ASCO, the world's leading professional organization for doctors that care for people with cancer.
In today's podcast, we'll discuss some of the research on gastrointestinal, or GI cancers, presented at ASCO's 2015 Annual Meeting. This podcast will be led by Dr. Jeffrey Meyerhardt, who is the clinical director and senior physician at the Gastrointestinal Cancer Center at the Dana-Farber Cancer Institute, and associate professor of medicine at Harvard Medical School. Dr. Meyerhardt is also an associate editor for Cancer.Net.
The ASCO Annual Meeting is the premiere educational and scientific event where physicians, researchers, and other healthcare professionals gather to discuss the latest in cancer care and treatment. This Cancer.Net podcast helps put new research findings into context and explains what they mean for patients. ASCO would like to thank Dr. Meyerhardt for summarizing this research.
Dr. Meyerhardt: Hello. My name is Jeffrey Meyerhardt. I'm a gastrointestinal oncologist at the Dana-Farber Cancer Institute in Boston, and I'm going to report on several of the abstracts that were presented at this year's ASCO meeting in Chicago. I'm going to cover gastrointestinal oncology abstracts and I'm going to divide them into abstracts related to colorectal cancer, and then some abstracts for cancers of the GI track that are not part of the colon and rectum, which is how ASCO divides their presentations. So let's start with just some abstracts that were presented in the colorectal oral session at ASCO this year. So the first two abstracts I'm going to talk about relate to patients who have unresectable—meaning that can't be removed—metastatic spread, colorectal cancer, and whether or not what certain—what are called localized therapies, therapies directed specifically at individual metastasis add benefit above chemotherapy alone.
And there were two very interesting abstracts presented at this year's meeting. The first one was presented by Dr. Theo Ross from the EORTC Cooperative Group. That's the European Cooperative Group that is a group of different cancer centers and cancer practices that joined together to do various clinical trials. And what they were trying to look at was the role of radiofrequency ablation in patients who had colorectal cancer that spread to the liver, but was not considered resectable. Radiofrequency ablation is a technique and a technology that has been around for quite a few years. It's basically going directly into a metastasis and creating radiofrequency waves that causes a heat reaction to try to kill the cancer cells.
There's various types of ablations: there's cryoablation which is freezing, radiofrequency ablation which is heating. This will focus on radiofrequency ablation. And what they did is they studied 152 patients, again, with unresectable liver metastases colorectal cancer. Patients had to have less than ten metastases, and none of the metastases could be greater than four centimeters because radiofrequency ablation does not work as well in ablations that are greater than four centimeters. And in this study, they could not have disease outside their liver. And patients were randomized, 152 of them, to receive radiofrequency ablation with chemotherapy, versus chemotherapy alone. The group did show that the patients that were randomized to include radiofrequency ablation had an improvement, what's called progression-free survival time, until their disease grew. And overall, survival in both of them were about a 40% improvement, so there was some improvement with adding radiofrequency ablation.
The caveats of this study—it was a small sample size, only 152 patients—they actually were trying to accrue more patients, and those had difficulties in accruing because I think certain providers, they were doing localized therapies and other providers favored to stay with systemic therapy alone. So they had problems with accrual, and actually, even though the progress-free survival was improved with radiofrequency ablation, there was a higher instance of metastases outside the liver that incurred in the group that received radiofrequency ablation. So I think these are data that fortunately is randomized to be able to look at technology that's out there. I think we need more data to really know which patients to use radiofrequency ablation and how the timing of it related to chemotherapy, but it's good that we have some data to start looking at this question.
Another localized technique was presented by Peter Gibbs and colleagues from the Royal Melbourne Hospital in Australia. They were looking at a technology called selective internal radiation therapy, or SIRT, and that uses yttrium-90 microspheres. Basically these are little tiny spheres that deliver radiation, again, to a localized area, and in this case, in liver metastases. It's delivered through the hepatic artery, one of the large blood vessels that's in the liver, that is actually the one that tumors try to get a blood supply from in a single delivery. They looked at 530 patients who were not surgically resectable for liver metastases. In this study, they were allowed to have disease outside the liver, but very limited. They could have a few lung metastases, they could have a few lymph nodes, but not extensive disease outside the liver.
The overall goal of the study was to look at progression-free survival, and the study actually was negative. The patients who were randomized to receive SIRT therapy with chemotherapy did not, overall, have a better progression-free survival than compared to the patients who had chemotherapy alone. If you look just at the liver, there was about a 37% improvement in progression-free survival, but again the overall progression-free survival when you looked at disease also outside the liver was not different. They have not reported yet on overall survival and there's actually two other trials that are looking at this technology as well, and the plan is to pool all three trials together to look at overall survival. So it's going to be a few years until we really know what the role of SIRT therapy is adding to patients with metastatic colorectal cancer. But again, I commend the authors for looking at this question to randomize study to really understand how these technologies can be integrated into the care of patients with metastatic colorectal cancer. The other caveat to consider is these patients are very variable. Some people have one liver metastasis, some people have multiple, some people have disease outside, there's some who don't. It can be hard to fully understand, from this trial, how to sort out which is the appropriate population, but again it's a good start to be able to understand these technologies.
We then had two abstracts that looked at, what some people would describe as, alternate or non-traditional chemotherapy interventions in patients with colorectal cancer. The first abstract was by Kimmie Ng and colleagues from my institute, the Dana-Farber Cancer, along with Alliance Cooperative Group. It's in National Cancer Institute, NCI Cooperative Group which are, again, group of cancer institutes as well as community practices that work on various clinical trials. Dr. Ng's abstract focused on vitamin D in patients with metastatic colorectal cancer. She is the cohort of group of patients from a large clinical trial called CALGB 80405, which was a clinical trial that was presented at last year's ASCO, where patients with metastatic colorectal cancer who had not had prior chemotherapy were randomized to chemotherapy with either a drug called bevacizumab, an antiangiogenic inhibitor, or a cetuximab, an EGFR inhibitor. And the chemotherapy could either be a combination of five if you look for an oxaliplatin called FOLFOX, or five if you look for an irinotecan called FOLFIRI. So it was a randomized trial looking at two different biological therapies added to standard cytotoxic chemotherapy. That trial showed no difference between bevacizumab and cetuximab. But Dr. Ng looked at the patients who enrolled in that trial where they had a measure of vitamin D level at baseline in their blood. And what she showed is a 35% improvement in overall survival for those who had the highest level of vitamin D, compared to the lowest level was a significant trend with increasing levels of serum vitamin D. She also showed a 21% improvement in progression-free survival for those with the highest level of vitamin D compared to the lowest level, again, with a significant trend with increasing levels.
Now, the caveats of the study is what's called an observational study. So Dr. Ng basically measured blood levels at a single time point in the patients when they were first starting treatment for metastatic colorectal cancer. There was no intervention. There was no attempt to try to change their vitamin D levels by supplementation with vitamin D. And so whether or not intervening to increase these patients' level would lead to an improved outcome in those who had the lowest levels is just not known yet, though there are now studies underway to try to look at this question.
A second abstract by Simer Bains and colleagues from the University of Oslo focused on aspirin in patients with colorectal cancer. Dr. Bains and her colleagues linked a cancer registry in Norway with a prescription drug database, in which they were able to find 25,000 patients who had colorectal cancer, and then they were able to look at what various prescription drugs those patients took. And she found 6,000 of those 25,000 colorectal patients took aspirin. And it lands up in Norway those who'd had aspirin mainly got it through a prescription drug plan as opposed to over the counter, which is more common in the United States. And what Dr. Bains showed is those taking aspirin had a 15% improvement in overall survival and a 25% improvement in colorectal cancer specific survival.
Now, the caveats of the study, again, is not randomized, it's, again, an observational study. So it was looking at what people were taking who all have had cancer, who were getting other standard treatments for cancer, it's possible some of them were taking non-prescription aspirin, not measured. So this could also affect the data. And also she's not able to measure the patients' compliance. So just because you pick up a prescription for aspirin doesn't necessarily mean you're taking it. But again, she does show a statistically significant improvement. And these data are actually consist of multiple other observational studies that shows that patients particularly with non-metastatic colorectal cancer early stage disease who take an aspirin, they have a lower risk of recurrence, lower risk of subsequent polyps in their colon, and we are currently now engaged in a large randomized study through the NCI Cooperative Group, looking at an inhibitor like aspirin called celecoxib, which are called a COX-2 inhibitor. And the patients who have stage three colorectal cancer are being randomized to celecoxib versus placebo for three years, in addition to the standard chemotherapy given for stage three patients. That study is still ongoing in accrual, and will be several years until we know the results of the study, but we're now having randomized studies to try to look at this very important question.
Two other abstracts that I'm going to point out is related to colorectal cancer, start looking at specific pathways that may be specific to certain people's metastatic colorectal cancer. So the first one I'm going to talk about is a paper by Salvatore Siena and colleagues from Italy. Dr. Siena and his colleagues looked at patients who had metastatic colorectal cancer who had an overexpression of a particular protein called HER2. HER2 is a protein that's also overexpressed in certain breast cancers and has been a very successful strategy for women who have HER2-positive breast cancers, using a drug called trastuzumab initially, now several other agents that inhibit that pathway, in addition to other standard therapies.
In this trial for patients with metastatic colorectal cancer, Dr. Siena's group looked at a dual inhibition of the HER2 pathway using two different drugs: one drug called trastuzumab, which is an antibody against HER2, and another drug called lapatinib, which is a pill form of an inhibitor inside the receptor for this pathway, to try to block the HER2 pathway. Patients with metastatic colorectal cancer who progress on all other standard therapies, who were considered HER2-positive, also have a wild type of a different protein called KRAS were screened. In total they screened about 850 patients, and only a very small percent had overexpression of HER2, only 5.4%. And of those 46 patients, 24 enrolled in this trial in which they all received trastuzumab and lapatinib. So again, it's a very small percentage of the overall colorectal cancer patients, but of those 24 patients who enrolled, 35% had some radiographic response, with one of them having a complete response, and seven of them having partial response. So again, these are all patients who also had multiple other chemotherapies before.
For those who responded, the time to progression was 5.5 months, and it did seem to relate to the level expression. So it was longer in those who had higher-level expression. The caveats, this is a very small sample so far, it's also a very small percentage of the patients with colorectal cancer, but it is data that has intrigued a lot of those in this field. And there are other trials now underway being developed to be able to look specifically at this small slice of the colorectal cancer, both to see if this data holds up, as well as to see should some of these drugs be moved up earlier in the treatment of patients with colorectal cancer who have overexpressed HER2.
The last paper I'm going to talk about, related to colorectal cancer looks at, what I call checkpoint inhibitors or immunotherapy, which is a topic that has really received a lot of attention in oncology in the last several years. And there are multiple cancers that seem to have successful treatment with the immunotherapy and the checkpoint inhibitors, and this is looking specifically at a subset of those who have colorectal cancer. So the abstract is by Dr. [?] and colleagues. They looked specifically at a type of colorectal cancer called the mismatch repair-deficient tumors. When DNA divides, sometimes it makes mistakes, and there are proteins that try to correct those mistakes. If those proteins are not working, these can end up having some mistakes that eventually could be mistakes in critical areas in the genetic footprint of our cells. And so, if these proteins are in error, they're called mismatch repair. And so it ends up some colorectal cancers that develop, have mismatch repair-deficiencies, and it's about 15% overall. Some of them are related to genetic syndrome called Lynch syndrome, and some of them are sporadic, i.e. that you don't necessarily have a genetic syndrome, but it happens just within the tumor.
And so this abstract looks at mismatch repair-deficient tumors. It took 25 mismatch-deficient colorectal cancers and actually 21 mismatch repair-deficient other types of cancers - so endometrial, ovarian and a few others. And they also looked at colorectal cancers that didn't have a deficiency in this pathway. So 25 mismatch repair-proficient colorectal cancers. So a total of almost 75 patients were analyzed in the study. And the study was looking at a pathway called the PD-1 and PD-L1 pathway. So it lands up, these are proteins that bind to each other, one from the tumor and one from the body's immune system, to try to prevent the immune system from fighting against the cancer. And so binding a PD-1 and PD-L1 inhibits, what are called effector T cell function, and there's various inhibitors that are being developed right now by various companies. One of the one that's further along in development is an antibody called pembrolizumab. It binds to PD-1 that prevents the binding of PD-1 to PD-L1.
So in this study, which had a goal of 75 patients overall, they presented an interim analysis, so the trial is not fully enrolled yet, but they've already enrolled 13 patients who had colorectal cancer with mismatch repair-deficient tumors. And of those 13 patients, 62% of patients who had a significant radiographic response, and 92% had some level of disease control, which is a really high level for a population that had multiple prior therapies that they had progressed on. Interestingly, the 25 mismatch repair-proficient colorectal cancer - so those that did not have a defect of the mismatch repair genes in colorectal cancer - none of them had a radiographic response. And only about 15% of them had disease control. So there's clearly a difference between the 13 patients who have deficiency in the mismatch repair, versus those who have proficiency in terms of activity. And interestingly, the ten tumors so far that have been enrolled with mismatch repair-deficiency, that are not colorectal cancer, also had a fair amount of activity with pembrolizumab - 60% response rate, and about 70% disease control. So there really seems to be an actor drug in a very small number of patients thus far, related to this mismatch repair concept. And so the caveat of this study is it's still a small number of patients. So they're still enrolling to try to get 25 mismatch repair-deficient colorectal cancers to see if this holds up, as well as mismatch repair other cancer types. And there are now efforts being considered how to test this in a larger setting, to see if these data are truly robust.
I'm now going to focus on some of the abstracts, and what was called the non-colorectal cancer sessions. In addition to colorectal cancer, the other gastrointestinal oncology cancers include esophagus, stomach, pancreas, liver, neuroendocrine, and carcinoid tumors. And I'm going to focus on a few of the key ones that came out from this year's ASCO. So I'm going to start by looking at an abstract related to gastric cancer. There was an abstract presented by Young Joo Bong and colleagues, with an analysis from what was called the KEYNOTE-012 trial. And do again, this is staying on the topic related to the checkpoint inhibitors and immunotherapy. This was a trial of patients with recurrent or metastatic gastric cancer, or gastroesophageal junction tumors, who were PD-L1-positive. Again using this PD-1 and PD-L1 inhibitor therapies, these are patients who are PD-L1-positive and they were treated with pembrolizumab.
This data was initially presented at ESMO, which is the European Society of Medical Oncology, back in the fall of 2014. This is an update of the data, in which they looked a little bit about the expression of the proteins in cells. There were a total of 39 patients enrolled in this trial, so it was a very small trial, and they were patients who had tumors that were already treated with multiple other chemotherapies. Of those 39 patients, there was a 22% response rate, and an over 50% disease control. And of those patients who had a response, the medium duration - how long it lasted - was about 40 weeks. It did seem to be active in this population of patients with gastric cancer. It was a manageable safety profile in those patients. And this particular aspect looked at the levels of PD-1 and PD-L1, and response to this, and showed that those who had higher levels seemed to have more increased activity. The caveats again is this is a small study of patients - only 39 patients. But it was interesting enough of a study that various companies, including the company that makes pembrolizumab, has mounted efforts to try to test this in a larger setting. And so there's currently ongoing trial looking at these drugs in gastric and gastroesophageal junction tumors. And so we'll know if this activity truly holds up in a larger population of patients.
The next two abstracts I'm going to talk about relate to carcinoid and neuroendocrine tumors—a very rare cancer relatively, but a cancer that has now had some increasing attention in the last several years. Because previously, it was a cancer that there were really no treatments for, and now, fortunately, we do have some treatments that help these patients. There are cancers that have very variable disease course, there could be cancers that can take years and years to really see growth and there are cancers that are carcinoids and neuroendocrine - they're more aggressive in nature.
There were two abstracts presented, both from clinical trials that came out of the NCI Cooperative Groups. The first trial that I'm going to talk about is one that was led by James Yao from MD Anderson and his colleagues. It was patients who had a carcinoid tumor, either in the colon, rectum or stomach. And they were randomized to bevacizumab, an angiogenic inhibitor with octreotide, versus interferon alfa-2b in octreotide. This trial was started thinking that bevacizumab would have activities to the disease, and interferon, there was some data-suggestive activity but it wasn't clear how active it was. And so 402 patients were enrolled and they actually showed no difference between these two arms. They both had activity, there was some higher response rate in the patients who received bevacizumab and longer time to treatment failure. And most of the reason for treatment failure was actually because of the toxicity of Interferon.
Interferon is a SubQ injection, so subcutaneously administer the patient several days a week. It can cause a fair amount of fatigue, some diarrhea and flu-like symptoms, and so patients have a tougher time tolerating Interferon, but it actually does have some activity in this disease, and so both of them seem to be active drugs. In fact, the interferon was more active than predicted, which is one of the reasons that it was thought that there was no difference between the two arms in this trial.
The second abstract was by Matt Kulke from the Dana–Farber Cancer Institute and his colleague - again an NCI Cooperative Group trial - in which he looked at neuroendocrine tumors that start in the pancreas of patients with advanced pancreatic neuroendocrine tumors. And they were randomized to a drug called everolimus. First is everolimus and bevacizumab. So everolimus is already FDA-approved for neuroendocrine tumors of the pancreas. This was looking at, was there additional benefit of adding an angiogenic inhibitor, bevacizumab, to everolimus? 150 patients were randomized, half of them receiving everolimus alone, half of them with everolimus and bevacizumab. And progression-free survival is not statistically significant. So this did not reach its primary endpoint in the study. It did favor bevacizumab, but it was not statistically significant and there was no difference in overall survival. There was some increase in toxicities in the arm that received bevacizumab, including hypertension, protein in the urine, diarrhea and cardiac events. So at this point, based on this trial, bevacizumab doesn't seem to be clearly adding benefit. But there was going to be additional analysis in the study to see if there's certain populations within that group of patients with neuroendocrine pancreas tumors that may receive a benefit.
Just focusing a little bit on neuroendocrine and carcinoid tumors now. This is, again, a cancer-type that had really no treatment options until last several years. And now based on several trials, there are various drugs that seem to have activity. Octreotide showed activity in a trial called the PROMID trial. Lanreotide has recently been FDA-approved based on a trial called the CLARINET trial. Both of them are drugs - one given intramuscularly, one given subcutaneously - that will give long-acting octreotide to patients that seem to have a benefit in disease control as well as symptom control that can be associated with neuroendocrine in carcinoid tumors. Sunitinib was FDA-approved for advanced pancreatic neuroendocrine tumor. And I guess as I said earlier, everolimus is FDA-approved for advanced pancreatic neuroendocrine tumors. And several weeks ago, there was a trial called the RADIANT-4 trial that suggested activity in carcinoid tumors as well, though we need to see that data further to understand how it is going to fit in with the treatment of carcinoid tumors.
Now lastly, I'm going to talk about two trials related to pancreatic cancer. Both of the trials were in an earlier setting, so patients not necessarily of metastatic pancreatic cancer, but either with resectable or potentially resectable pancreatic cancer. So the first abstract was by Marion Sin and colleagues from Germany. It was called the CONKO-005 Trial. The CONKO group is a group, again from Germany, that has looked at various treatment strategies for patients with resectable pancreatic cancer. It is the group that found that gemcitabine, given as an adjuvant setting after the resection of pancreatic cancer, shows a benefit compared to surgery alone. The CONKO-005 trial took 436 patients who had resection of pancreatic adenocarcinoma cancers and they were randomized to gemcitabine, the standard that was previously set, versus gemcitabine plus a drug called erlotinib, an inhibitor of the epidermal growth factor receptor, and it's in a pill form that's FDA-approved for lung cancer and has shown a benefit in lung cancer. And it's actually FDA-approved for metastatic pancreatic cancer based on a trial that showed really a marginal increased benefit over gemcitabine alone. But, it was again tested in metastatic pancreatic cancer, not in patients who had resected pancreatic cancer. Patients were randomized again to gemcitabine versus gemcitabine and erlotinib. They were treated for 24 weeks, so about six months, and the primary endpoint was disease-free survival. Unfortunately, it did not show any additional benefits to disease-free survival. So adding erlotinib did not add benefit above gemcitabine alone.
The last abstract that I want to talk about is an abstract by Matt Kass and colleagues, again using a clinical trial through the Alliance Cooperative Group. This was actually a very small trial, but I'll bring it up, and it was selected as an oral presentation because it really tested a concept that is increasingly being considered in the treatment of patients who have potentially resectable pancreatic cancer. It was a pilot study of patients who had borderline resectable pancreatic cancer. So it lands up some patients when they're newly diagnosed with pancreatic cancer. It may not have evidence it's spread outside the pancreas, but in the pancreas, may be touching or partially surrounding certain blood vessels, which is generally considered a non-resectable disease at that point. And the question is, do we have strategies to try to pull that tumor away from those vessels to be able to potentially make it resectable? With more active chemotherapy now used for metastatic pancreatic cancer, the question is, can those therapies be used in this setting to potentially make patients resectable?
So the treatment strategy for this was to give patients a combination called FOLFIRINOX - 5-FU, leucovorin, already taken Nexalin plan, which is a combination now used in metastatic pancreatic cancer patients. They've treated them for several months with FOLFIRINOX. They then repeated staging scans. If again the patient did not have evidence of metastatic disease, they were then treated with radiation, along with a drug capecitabine, an oral form of 5-FU. They were then restaged. And again, if they didn't have evidence of metastatic disease and potentially appeared resectable, they underwent surgery. And then after surgery, they could have two more months of gemcitabine.
It was only 22 patients tested. But again, it was a pilot study to see if the strategy could be successful. Of those 22 patients, 21 landed up going forward to be able to get radiation. So they got FOLFIRINOX, they got restaged, and then they got radiation. 15 of those 22 landed up having a pancreas resection. Ten landed up getting adjuvant gemcitabine in addition. And so nine overall went through all the prescribed protocol therapy. But really, probably the most interesting number that came out of this trial, again these are patients who were considered borderline resectable, where you weren't sure, could you cut out the tumor or not. And 93% who underwent resection, landed up having what's called an R0 resection. So that means they were able to get negative margins. There was no tumor at the edge of the sample that was sent to the pathologist. So 93% R0 resection for those who had surgery, and so that again is an impressive number. It was 15 of the 22 underwent resection and really the vast majority of them had a successful surgery. Now, how that's going affect overall their outcome and their recurrence, we don't yet from the data from this trial. This small pilot study certainly supports trying to do a larger effort to know how to sequence these therapies in patients who are potentially resectable, or borderline resectable pancreatic cancer to try to maximize the outcome in this disease.
So those are the highlighted abstracts that came out of the ASCO meeting. There are certainly lots of work still to be done on multiple of these efforts, but they're efforts that really are starting to look at different strategies for a lot of these diseases, and we look forward to trying to learn more on how to capitalize and to better treat multiple of these cancers. Again, my name is Jeffery Meyerhardt from the Dana-Farber Cancer Institute, and thank you for listening to this podcast.
ASCO : Thank you, Dr. Meyerhardt. To find all of the science presented at ASCO's 2015 Annual Meeting, visit www.cancer.net. If you have questions about whether new research may affect your care, be sure to talk with your doctor.
Cancer.Net is supported by the Conquer Cancer Foundation, which is working to create a world free from the fear of cancer by funding breakthrough research, sharing knowledge with physicians and patients worldwide, and supporting initiatives to ensure that all people have access to high-quality cancer care. Thank you for listening to this Cancer.Net podcast.