June 25, 2015

expert reaction to possible new biomarker for early-stage pancreatic cancer

A new study published in the journal Nature has reported the identification of a blood protein that has the potential to be used as a biomarker for early-stage pancreatic cancer, which the authors hope may be able to detect the disease before current standard tests.

Prof. Dorothy Bennett, Head of the Molecular Cell Sciences Research Centre, St George’s, University of London (SGUL), said:

“Pancreatic cancer is the tenth commonest cancer overall in the UK (excluding non-melanoma skin cancer), but the fifth highest cause of cancer death. It causes a disproportionate amount of mortality because it tends to be diagnosed late, after it has already spread and cannot be cured by surgery.

“That makes it potentially very interesting that the exosomes [tiny bubbles of cell membrane and cytoplasm] in the blood serum of both patients and mice with pancreatic cancer or benign pancreatic diseases, are here reported very consistently to show high levels of GPC1 only when the patients or mice did have cancer, and with a striking 100% accuracy, i.e. no false positives and no false negatives.

“This would be very good news for patients suspected of having this cancer.  The methodology looks sound; all assays were done blind and sample identities decoded afterwards.  Sample sizes and replications look good.  Even if any false positives emerge later, a test with “nearly” 100% accuracy would still be valuable and desired.

“This study strongly suggests that a way to create a test for pancreatic cancer has been found in principle.  The technology used here is quite widely available, however further work will be needed to develop an antibody that can be sustainably used in commercial tests. This research will need further funding from a company before a standardised and validated test can be routinely used in hospital labs. Overall, this would likely take a few years to develop.

“The authors also found this GPC1 marker on exosomes from other kinds of cancer cell lines (including colon, breast, melanoma).  So if such a test were to be used for screening people in general, it is possible that it might pick up cancers of different kinds, not just pancreatic.  More experiments would be needed to find out how many types of cancer might give a positive test, and the authors are currently talking only about pancreatic cancer.

“The test would not predict the course of the cancer beyond simply detecting it and giving an estimate of its current size. It is just a snapshot at one time.  It may well predict in the sense of picking up very small cancers that will become detectable only later by other means.  But I can’t see a way that it would predict whether a cancer would progress more or less quickly from its current size compared to others of that size.”

Prof. Paul Pharoah, Professor of Cancer Epidemiology, University of Cambridge, said:

“Pancreas cancer is one of the cancer for which there has been little improvement in survival in the past 30 years.  One of the reasons for this is that the disease is often diagnosed when it is advanced and it cannot be cured by surgery with or without chemotherapy.

“This paper reports on a newly identified biomarker – known as GPC1 positive exosomes – that might be used to identify pancreatic cancer at a sufficiently early stage of the disease to improve outcomes.  While the findings are interesting and the potential is clear they are in themselves very preliminary and much work needs to be done to show that GPC1 positive exosomes really do have clinical usefulness.

“A key requirement for a biomarker like this is that is can detect the disease of interest in the early stages before clinical symptoms are apparent -i.e  when the cancer is small and before it has spread.  In this paper, the authors suggest that this will be true, but they only studied blood samples from 5 patients with pancreas cancer precursor lesions.  Apart from the very small sample size – too small to draw robust statistical conclusions – it is also notable that while these 5 patients all had precursor lesions they also all had symptoms and had already had their pancreas disease diagnosed by other means.  In other words the GPC1 positive exosomes were shown to be present in symptomatic patients, but there is no evidence that the GPC1 positive exosomes would be present before symptoms develop.  In addition no information is given about the timing relating to when samples were taken.  It is plausible that the investigation of these patients with symptomatic pre-cursor lesions has led to the GPC1 positive exosomes being detectable in the blood.

“The second part of the study reports that a reduction in the levels of GPC1 positives exosomes after surgery compared to the level before surgery is predictive of outcome.  This, in itself is merely an interesting observation and further work is also needed to establish whether this is a meaningful finding.  For example, the post-surgical level may simply represent how much of the tumour remains after surgery.  Much larger studies will be needed to demonstrate that GPC1 positive exosome levels are a better predictor than other important prognostic factors.

“In summary, this paper reports some intriguing findings, but they are preliminary and much more data will be needed to demonstrate that GPC1 positive exosomes can be used as a useful biomarker for the early detection of pancreas cancer or for guiding treatment in patients with pancreas cancer.”

 Prof. Eric O’Neill, Associate Professor of Radiation Oncology, University of Oxford, said:

“Currently, the majority of pancreatic cancers are diagnosed at late stages of disease – making interventions less likely to be successfully due to metastatic spread. Early lesions that are picked up through routine scans offer more treatment options including surgery, however, as pancreatectomy is itself debilitating, we need to ensure that pre-neoplastic lesions actually pose a significant health risk and are not benign or cyst tissue. The current study reports the ability to differentiate benign from early cancer by monitoring blood levels of cancer cell derived microvesicles termed exosomes that are released into the circulation. Moreover this ‘smoking gun’ of cancer cell activity, highlights the presence of these cells at microscopic levels that are currently below detection limits of medical scans. The identification of a single biomarker that can both detect the presence of cancer cells and define their potential aggressiveness, i.e. requiring intervention, will undoubted increase patients that can be provided with clinical strategies from which they are more likely to derive significant benefit and which, in turn, should increase survival rates.

“Although the paper appears as a robust clinical validated analysis, the Glypican-1 molecule is implicated in a number of cancers, including breast. Early detection will therefore highlight potential for disease but not necessarily specify pancreatic cancer, resulting in complications on a potential course of actions following a positive result. Is may too early to say this represents a reliable test for pancreatic cancer as we first need to see the results replicated in other centres and determine what the threshold is for normal or non-cancer detection, but the signs are extremely encouraging and in line with an emerging view that exosomes may represent the best and most reliable way to detect and track disease. As the authors clearly demonstrate the levels of exosomes do track with disease burden in animals, the first clinical utility is likely to be in tracking minimal residual disease and the appropriate timing of invention schedules post-surgery to prevent or extend the period to relapse.”

Prof. Alastair Watson, Professor of Translational Medicine, University of East Anglia (UEA), said:

“A reliable blood test for pancreatic cancer is a holy grail in cancer medicine. Pancreatic cancers are usually diagnosed after when curative treatment is impossible. As a result most pancreatic cancers are fatal.

“All cells release small packets of proteins and nucleic acids called exosomes. Professor Kalluri and his team have discovered that exosomes from pancreatic cancers contain a protein called Glypican-1. Exosomes from normal cells do not contain this protein. They show that detection of exosome containing this protein enables the detection of pancreatic cancer at a much earlier stage than is currently possible raising the prospective that surgery will be curative. Their experiments suggest that this new blood test can detect pancreatic cancer before they are visible on MR scanning. Current blood tests are not accurate enough for the diagnosis of early pancreatic cancer. The new blood test is a significant advance on current diagnostic tests. Unfortunately the new test is very complex to perform and many routine diagnostic labs will not have the equipment required. Further development work will be required before it can be brought into hospital laboratories.”

‘Glypican-1 identifies cancer exosomes and detects early pancreatic cancer’ by Melo et al. published in Nature on Wednesday 24th June. 

Declared interests

None declared.