Can protein science solve the unmet needs in pancreatic cancer diagnosis and therapy?

KEYWORDS:

• Biomarkers
• diagnosis
• health economy
• pancreatic cancer
• prognosis
• proteomics
• treatment

Pancreatic ductal adenocarcinoma is the most common subtype of pancreatic cancer. The disease is associated with a grave prognosis and no major advancements concerning treatment have been made during the past several decades [1]. Currently, pancreatic cancer is the fourth most common cause of cancer-related death, but is predicted to become the second leading cause of cancer death by 2020 [2].

Pancreatic cancer pathogenesis is complex. The recently held belief that pancreatic cancer is a slowly progressing disease, with precursor lesions progressing to invasive disease over many years, has been challenged [3]. It appears that at least some pancreatic tumors do not follow the stepwise progression model and instead develop rapidly and metastasize early. Even small tumors, 5 mm in size, can be associated with distant metastases and poor prognosis [4].

The societal burden of pancreatic cancer is substantial. The health-care costs associated with pancreatic cancer has been calculated to be approximately 86–93 million euros annually in Sweden [5]. Furthermore, pancreatic cancer is reported to be the fifth most costly cancer site at an expense of 3.9 billion euros annually in Europe [6]. Screening programs for pancreatic cancer do exist in patients with hereditary risk factors and known genetic syndromes associated with pancreatic cancer, using conventional imaging techniques such as magnetic resonance imaging and endoscopic ultrasound. However, the evolution of proteomics and less invasive screening modalities using protein biomarkers has been shown to be cost-efficient. Calculations were made on a population consisting of elderly patients with newly onset diabetes mellitus, where the pancreatic cancer incidence was 0.71%. In this proposed setting, the cost of a protein biomarker-based test was 400 euros for a test with a sensitivity of 88% and a specificity of 85% [7].

Because of the lack of early clinical symptoms and screening biomarkers, much effort is put into discovering novel biomarkers. In 2009, there were more than 2500 biomarkers suggested for pancreatic cancer [8]. However, none of these biomarkers have been validated for clinical use. CA 19-9, though insufficient, continues to be the golden standard. The European Group on Tumor Markers report sensitivity and specificity values of 81% and 90%, respectively, for CA 19-9 [9]. Unfortunately, many issues are associated with CA 19-9. It is elevated in patients with obstructive jaundice [10] and approximately 5% of the population is negative for the Lewis blood antigen and cannot produce Lewis A antigen [11]. These limitations, combined with the relatively low sensitivity and specificity values, make CA 19-9 suboptimal for diagnostic screening. It is preferably used to monitor response to cancer therapy [12].

Traditionally, antibody-based techniques such as ELISA or immunohistochemistry have been the major methods used for targeted protein quantification and validation. However, lack of specific antibodies remains a challenge. The recent developments of mass spectrometry-based platforms have revolutionized quantitative proteomics enabling the systematic and quantitative assessment of disease-perturbed proteins in various biological samples [13]. Mass spectrometry can be applied to the discovery of novel biomarkers for pancreatic cancer. Designated biomarker candidates can then undergo targeted mass spectrometry-based verification and validation with absolute quantification using multiple or selected reaction monitoring [14]. Major directions in pancreatic cancer proteomic studies include the search for biomarkers for early detection and biomarkers that can guide prognosis and predict response to chemotherapy.

It is reasonable to believe that a panel of biomarkers may be required to achieve high enough sensitivity and specificity for pancreatic cancer diagnosis. Protein science is, however, not only useful in the detection of pancreatic cancer. The mapping of proteins expressed by individual pancreatic cancers could be used in predicting response to therapy. This information could be used to personalize therapy given to patients to achieve the best care possible and to avoid unnecessary treatment.

Gemcitabine is the standard chemotherapy used for the treatment of pancreatic cancer, in the adjuvant as well as palliative setting. Human equilibrative nucleoside transporter-1 (hENT 1) is a transmembrane protein that mediates cellular entry of nucleosides and nucleoside analogs such as gemcitabine. hENT1 is a possible biomarker for assessing response to gemcitabine treatment [15]. Selecting patients for gemcitabine treatment based on hENT1 status may be cost-saving for the society [16].

In the metastatic setting, combination chemotherapies, including FOLFIRINOX and gemcitabine/nab-paclitaxel, have been found to improve overall survival rates [17,18]. Biomarkers for predicting the response to these treatments could be searched amongst proteins related to drug metabolism and activity of the individual cytotoxic components. FOLFIRINOX is a combination of the drugs 5-FU, leucovorin, irinotecan, and oxaliplatin. Possible biomarkers include dihydropyrimidine dehydrogenase and thymidylate synthase, which are enzymes involved in 5-FU metabolism and activity [19,20]. Carboxylesterase-2 is a protein essential for converting irinotecan into its active form SN-38 [21]. Markers of oxaliplatin response are less characterized, but it has been found that RABL6A, a novel RAB-like protein, promotes oxaliplatin resistance in pancreatic tumor cells [22]. Secreted protein acidic and rich in cysteine (SPARC) was initially believed to be a predictive marker for nab-paclitaxel response, but this association could not be confirmed when a larger study was conducted [23]. Other potential biomarkers for nab-paclitaxel treatment include markers predictive of taxane resistance such as tubulin expression (e.g. overexpression of the βIII-tubulin isoform) and BRCA status [24].

In summary, pancreatic cancer continues to be a major challenge for the health-care system and the society at large. While some pancreatic tumors may develop progressively, at least some tumors do not follow the sequential progression model. This new information needs to be taken into consideration in the design of future screening and therapeutic approaches. Tailoring treatment to the individual patient’s phenotypic expression would ultimately realize the promise of personalized medicine. Proteomics is a field with much potential, but discoveries are yet to be translated to routine clinical practice.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Additional information

Funding

This paper was not funded.