Evaluation of Prostate-Specific Antigen Glycomics Assay for the Early Detection of Prostate Cancer

In the Netherlands, approximately 12,000 men are yearly diagnosed with prostate
cancer (PCa), making PCa the most frequent type of cancer in men. Elevated
prostate-specific antigen (PSA) levels (>3 ng/mL) are a first indicator of PCa.
Namely, a small amount of this protein can *leak* into the circulation, because
with PCa, the cells are misaligned causing elevated PSA concentrations (PSA
gets better access to the circulation). When an elevated level of PSA is found,
further examination is needed to determine the diagnosis (e.g. digital rectal
examination (DRE), prostate MRI) and prostate biopsies). Unfortunately, other
prostate related diseases (e.g. benign prostate hyperplasia or an inflammation)
can also result in elevated PSA levels. Furthermore, the PSA-test is unable to
differentiate aggressive from non-aggressive PCa. In the case of non-aggressive
PCa, the cancer is slowly growing and the chance that the patient will
encounter progression or any symptoms due to the cancer are rather small.
Taking this into account it is recommend to not immediately treat the patient,
but rather to start active surveillance (patient comes back on a regularly
basis and treatment will only be started when the cancer seems to progress).
With the abovementioned examples it is clear that there is a need for a better
test that is able to distinguish PCa from other prostate related diseases, to
avoid overdiagnosis and overtreatment of the patient.

The current research is focused on finding a method that is able to better
stratify the patients. A possible method could be to study PSA in more detail,
specifically the modifications which are present on the protein. One of these
modifications is glycosylation (sugars attached to the protein). These sugar
groups can be very diverse and recent studies have shown that changes in the
glycosylation can be related to various cancers and autoimmune diseases. The
purpose of this study is to examine whether changes in the glycosylation of PSA
can also be related to PCa. As certain changes might be able to distinguish
whether the patient has PCa and its aggressiveness. This knowledge would be a
huge benefit compared to the conventional PSA-test.
Recently, we have developed an in-depth PSA Glycomics Assay (PGA) that can
determine all these glycosylation features of PSA derived from urine.
Unfortunately, no direct correlations could be made between the clinical
diagnosis and the PSA-glycosylation features in urine. Our hypothesis is that
PSA from the circulation, but not from urine carries the cancer glycosylation
signature that can stratify PCa from other prostate related diseases and can
differentiate between aggressive and non-aggressive PCa. PSA that leaks into
the blood is expected to be largely tumour-derived, since only the tumour
region will show increased leakage of PSA into the circulation. We will compare
the glycosylation of the possibly tumour-derived plasma PSA (pPSA) with uPSA
that is derived from the entire prostate gland and presumably exhibits a
healthy-type glycosylation profile. We expect that this personalized diagnostic
approach that takes the individual glycosylation of a patient into account,
will result in improved diagnostic performance of the PGA.

In total, 225 patients are needed for this research, who will donate a blood
(15 mL) and urine (20 mL) sample.
It is estimated that 15% of the included patients samples will be incomplete
(either a blood or urine sample is missing), due to this 25 patients will not
be included in the research. Additionally, 50 patients are estimated to belong
to Group 2 (patients with elevated PSA concentrations but no indication for
biopts based upon DRE and MRI) and will also not be taken along in the research
question. From the remaining samples (Group 1) the protein PSA will be
immunocaptured and analyzed with capillary electrophoresis coupled to a mass
spectrometer at the LUMC. The samples will be anonymized and only the
participating urology departments will have access to the clinical details of
the patient (eg. Gleason score, cT stage, pT stage, prostate MRI results, etc).
This to avoid any bias that could be introduced while analyzing the
samples.After the analysis (100 samples per biofluid), the participants will be
divided into three groups after the analysis of the samples has been performed.
Group a will consist of the patients with elevated PSA concentration, but where
prostate cancer was excluded due to a negative biopsy. Group b will consist of
the patients with elevated PSA concentration where prostate cancer was found.
Group 2 will consist of patients with elevated PSA concentration (>3 ng/mL) but
no indication for prostate cancer based on MRI and DRE.
By making use of the abovementioned groups, it can be investigated whether the
PSA Glycomics Assay has an added value next to the current PSA-test. For each
sample a PSA glycosylation profile will be obtained and per patient their urine
profile will be first compared to their plasma profile. The difference in the
glycosylation profile will result in a delta per glycoform for each patient.
This will be used to eventually compare the observed delta*s per glycosylation
form between the different groups. Potential biomarkers will be compared to
current diagnostic tools such as histology, digital rectal examination and
prostate MRI to evaluate whether the found biomarkers are a true added value.
Validation will be performed in the case one or several biomarkers are found
with a new cohort consisting of 50 participants. During this phase the
specificity, sensitivity and positive predictive value will be determined. Next
to the evaluation whether the PSA Glycomics Assay can predict the presence of
prostate cancer it will also be evaluated if the assay can predict the level of
aggressiveness of the prostate cancer. For this purpose the Gleason scores
(obtained from histology after biopsy), TNM score and the PSA concentration
will be used. Patients with prostate cancer will be subdivided into three risk
groups: low-, medium- and high-risk.

The patient will be asked to donate urine (20 mL) and a blood sample (15 mL)
next to the regular examination (digital rectal examination, prostate MRI and
biopsy). This procedure will cost the patient some extra time. Every time when
blood is drawn a local bruising can occur. Where possible, the blood will be
drawn with a routine blood draw to minimize the risk. The patient will be asked
to collect his own urine sample, it is not expected there will be any
burden/risk due to this procedure.