Dutch and primarily other European researchers are about to begin a human clinical trial involving a regimen of two relatively new targeted therapies (afatinib plus selumetinib) for the treatment of pancreatic cancer. Afatinib is known as Gilotrif in the U.S., and is produced by the family-owned global but German-headquartered Boehringer Ingelheim Pharmaceuticals firm. Selumetinib is produced by the AstraZeneca firm (AZ); it is currently in the Phase III SELECT-1 clinical trial in combination with docetaxel for the second-line treatment of KRAS positive non-small cell lung cancer. AZ is partnering with Roche Molecular Systems to develop a test marker to help identify those patients who are most likely to respond to selumetinib.
The lead institution for the upcoming clinical trial involving pancreatic cancer will be the Antoni van Leeuwenhoek Hospital in Amsterdam. The work is based on earlier study by Bernards, Sun and other researchers who laid the predicate for this research in a recently published medical article on March 17, 2014 in the journal Cell Reports about the effects of this combination regimen in overcoming Kras resistant lung and colon cancer. The coming human clinical trial will also include lung and colon cancer, as well as pancreatic cancer.
The mutated ras oncogene is found in approximately 90% of cases of pancreatic cancer (pancreatic adenocarcinoma), and generally in about 20 – 30% of human cancer. These cancers tend to be ones that are particularly difficult to treat – with a tendency to poor outcome. In fact, one of the four specific initiatives proposed by the recent Scientific Framework for Pancreatic Ductal Adenocarcinoma of the U.S. National Cancer Institute is: “Developing new treatment approaches [for pancreatic cancer] that interfere with RAS oncogene-dependent signaling pathways.”
The recent study by Bernards et al. on Kras mutant colon and lung cancer found that suppression of the tyrosine kinase receptor ERBB3 had the effect of sensitizing these tumor cells to MEK inhibitors. Afatinib is a tyrosine kinase receptor inhibitor that not only affects epidermal growth factor receptor (EGFR – as does erlotinib -Tarceva), but also acts in a wider manner in these regards. Afatinib received U.S. FDA approval on July 12, 2013 for the treatment of metastatic non-small cell lung cancer which has EGFR exon 19 deletions or exon 21 (L858R) substitution mutations as detected by an FDA-approved test. Selumetinib blocks the enzyme MAPK kinase (MEK) that is just “downstream” from the BRAF gene and which is also involved in the Kras gene signaling pathways.
In the recent published study, the authors found that the combination of these targeted therapies acted synergistically in promoting cell death in Kras resistant lung and colon cancer cells, AND in significantly reducing tumor growth (to about zero growth over the 4-week study period) in referent mice models. It is easy to see why the authors have now added pancreatic cancer to the study.
Thus, the researchers have presented biological plausibility of the regimen in mutated Kras tumors and have established a basis for possible preliminary efficacy in pre-clinical studies. And so the process of human testing begins. It will be of some interest to see if the efficacy holds in human subjects, and what the tolerability profile will reveal. And the effect specifically on pancreatic cancer. This is interesting and promising work.
Dale O’Brien, MD
We speak of pancreatic cancer as if it is one thing, but there are many kinds of pancreatic cancer; one classification had this number at 17. A frequent anatomic / physiologic action of the pancreas is that it tends to produce cysts (and sometimes even “pseudocysts” – pockets of fluid often seen as a result of pancreatitis). Many of these cysts are benign, but some percentage are or become malignant. It is often difficult to differentiate between the innocent cystic structures, and ones that may become serious current or future problems.
Unexpected findings have become an increasingly frequent issue, as our ever advancing radiographic tools have progressed to a point where they often identify troubling elements in the pancreas (or elsewhere) which were not the target of the initial procedure – but now which we are must deal with in terms of their clinical significance – for the long-term good of the patient, and their peace of mind. The clever term has arisen for these unexpected findings: radiological “incidentalomas.” Pancreatic cysts then are often incidentalomas – requiring an element of diagnosis and explanation.
Also until now, carcinoembryonic antigen levels (CEA) – including the knowledge of its limitations – was often considered the general marker of choice for this task; a high level tending to point toward malignancy. We should also here begin to discuss individual “mucins” that are glycoproteins related to a series of a “family” of about twenty MUC genes. These mucins tend to line the surface of epithelial cells found in human lung, GI tract, eye, and other organs. Several of the MUC proteins have found to be overexpressed in colon, breast, ovarian, lung and pancreatic cancers.
Now comes a clever study by Jabbar and colleagues from University of Gothenburg, Sweden who published their results in the February 2014 issue of the Journal of the National Cancer Institute, outlining their work in utilizing mucin proteomics as a biomarker process to differentiate benign pancreatic cysts from those that may become precursors of pancreatic cancer.
Over about a five-year period of time (2007-2012) the authors evaluated 102 patients with what appeared to be pancreatic cysts, ultimately including 91 patients in their study who met their established inclusion criteria. The researchers then used endoscopic ultrasound to aspirate contents of the cysts, examining the extract for cytological (cell) changes, CEA levels, and individual MUC (mucin) levels including a panel of mucins. They found that MUC1, MUC2, and MUC5AC particularly appeared to identify cysts with potential for pancreatic cancer malignancy.
The results of this study were that mucin proteomic profiling (97.5%) was a more accurate method of identifying malignancy potential than either CEA levels (78.0%), or cytology (71.4%). The authors suggest that proteomic biomarkers involving a panel of mucin glycoproteins appear to be a promising manner of differentiating benign cycts from those with malignant potential for pancreatic cancer. This is a highly persuasive study that requires confirmation.
Dale O’Brien, MD
“It’s complicated.” But why should pancreatic cancer be any less convoluted than other aspects of the pancreas? The origin of the embryologic pancreas develops from TWO buds off of the foregut – that EACH “typically” moves and develops, eventually merging to form what we know as the pancreas. This somewhat individualistic merging process alone produces many anatomic anomalies, as any experienced pancreatic cancer surgeon will tell you. Also, separately, the function of the pancreas is complicated, divided roughly into exocrine (digestive juice producing) and endocrine (the hormone producing “islet”) areas.
Pancreatic cancer can arise literally from virtually any type of cell in the pancreas. The most common kind, that which has such a rough prognosis arises in the duct lining cells that line the pathways that carry digestive juice downstream (left to right) eventually usually combining with bile juice (at the bile duct junction) to deposit the combined product, so useful for ordinary digestion, into the first (duodenal) portion of the small bowel. This type of tumor then is called pancreatic ductal adenocarcinoma.
And what are we to make of the general observation of this relationship between this form of pancreatic cancer and what seems empirically to be an increased association of the onset of diabetes? After all, they are from completely different functional areas of the pancreas. Is this association real, or a form of recollection bias? If so, is diabetes mellitus a cause of ductal pancreatic cancer – or a result of it?
This is the vineyard in which Nikfarjam and his surgical colleagues at the University of Melbourne in Australia have been working in regard to their recent study published on March 9, 2014 in the Annals of Surgical Oncology. The researchers tried to better quantify the strength of the association between diabetes and pancreatic cancer over time, and to better qualify the relationship. After reviewing forty years of published medical articles as found in electronic databases (1973 through 2013), the authors identified eighty-eight studies that met their criteria. These included fifty cohort studies and thirty-three case-study studies involving the onset of diabetes and the risk of pancreatic cancer.
They found that the relative risk overall between the onset of newly diagnosed diabetes mellitus was statistically significant but at a reasonably modest 1.77. Though through a consideration of time as a factor, the authors conclusions demonstrated a more complex picture. They found that the closer in time to the diagnosis of diabetes: the more likely the discovery of pancreatic cancer. So, for example, the researchers calculated the relative risk at less than one year at 6.69 (the pancreatic cancer occurred nearly seven times more frequently than what one might expect), but that this factor dropped such that at ten years the relative risk was only 1.36.
This is a lovely study that further quantifies the elusive relationship between diabetes and ductal pancreatic cancer. There are still many questions, and it leaves the chicken versus egg question unsolved. But increasingly the evidence mounts that during the year or two or more after the onset of diabetes mellitus, a pancreatic cancer diagnosis is much more likely. That this is a high risk period that demands heightened clinical scrutiny.
Dale O’Brien, MD