Changes in Biomarkers From Blood Over Time in Patients With Pancreatic Adenocarcinoma

This is a single-institutional, prospective study at the Johns Hopkins Hospital which will be conducted from January 1, 2015 to December 31, 2018. All patients who present to the Johns Hopkins Hospital with a pancreatic adenocarcinoma (PDAC) for evaluation of treatment by chemotherapy, radiation therapy, and/or surgery will be approached to participate in this study, but only those who consent to the study will be included. All patients who consent to participate in this trial will have blood drawn (approximately 50 ml or 3-4 tablespoons) at the time of diagnosis but before any treatment had been administered. Further blood will be drawn after any treatment. For example, those patients that undergo surgical resection will also have blood drawn after resection prior to discharge. Those patients that undergo treatment with chemotherapy, radiation therapy, and or biologic/vaccine therapy will have blood drawn once after this treatment is completed. If a patient undergoes chemotherapy and then surgical resection, he or she will have blood drawn after chemotherapy but before surgical resection, and then again after surgical resection, all in addition to the pre-treatment blood sample (see schema included in IRB). For the first 2 years following surgery samples will be collected every 3-4 months. Beyond that the investigators will collect samples every 6 months for the next two years. For all patients found to be alive and disease free beyond 4 years after surgery samples will be collected once every year. This blood will be stored in the investigators' research facility and as such will be labeled with a de-identified code. If a patient refuses further blood draws at any time during the study or is loss to follow-up, the samples that have been collected will still be utilized.

The blood will be processed on the day of collection, within 6 hours of being drawn. Twenty (20) ml of blood, collected in two purple topped tubes, will be used for isolation of circulating tumor cells (CTCs) through the Isolation by Size of Epithelial Tumor Cells assay (ISET), which separates tumor cells by size onto a membrane which can be used for further experiments. These 20 ml of blood will be mixed with ISET buffer per standard protocol and filtered through the ISET machine onto membranes. These membranes will be stored at -20 degrees C under a de-identified number, and remain at these conditions for several years with full preservation of the cells. Further experiments will be done on these membranes, which include immunofluorescent staining, laser capture microscopy, and genetic analysis. The remaining 30 ml of blood, collected in red and purple topped tubes, will be processed by a standard protocol within 6 hours of collection. Blood will be double spun to remove white blood cells, and plasma and will be frozen. Driver gene mutations characterized in the resected tumors will be monitored by digital droplet Polymerase Chain Reaction (ddPCR), including the Kirsten rat sarcoma viral oncogene homolog (k-ras), p53, p16/cdkn2a and the Deleted in Pancreatic Cancer-4 gene (DPC4/SMAD4) to identify the presence of ctDNA and whether levels are rising.

Data will be collected, de-identified and analyzed as a single dataset at the Johns Hopkins Hospital. Data will be obtained from patient electronic and paper clinical charts from the John Hopkins Hospital starting on 1/1/2015, with only data in existence as of 1/1/2015 collected. More specifically, radiology, chemotherapy, operative and pathology reports, discharge summaries, and clinic notes will be reviewed and data from these sources collected. For patients lost to follow-up, the Social Security Death Index (but will not need to collect social security numbers) will be searched for evidence of death. Collected patient data will be entered into a dedicated database. Data will only be collected in order to correlate outcomes and tumor characteristics with the biomarkers identified during this study. Following data accrual and prior to analysis, all personal identifiers will be deleted and no identifiable information will be present at the time of statistical analysis. The subject data generated will not appear in the subject's medical record and will be used for research purposes only. The data will be recorded in a password-protected, secure database within the Principal Investigator's office and will be accessible only by the key personnel. After all data for a particular subject is collected, the subject's identity will be deleted to assure confidentiality.

The data that are collected in this study will not be used for the clinical management of any patient on study.

Inclusion/Exclusion Criteria

The investigators will include all patients who present to the Johns Hopkins Hospital for treatment with a diagnosis of a pancreatic adenocarcinoma. Subjects will only be included if they are over the age of 18 year old and plan to undergo their treatment at the Johns Hopkins Hospital. All patients with non-primary pancreas tumors, pancreatic neuroendocrine tumors, or benign pancreatic masses will be excluded. Patients will be excluded if they have already undergone treatment of any kind at another institution.

The investigators aim to include data from at least 200 patients over the three years of this study. The primary outcome will be the identify biomarkers in the blood from patients with pancreatic adenocarcinoma and the changes in these biomarkers over time in relation to treatment. These biomarkers will include specific antibodies, circulating tumor cells (CTCs) and circulating tumor DNA (ctDNA). Secondary outcomes will include the relationship of pre-treatment biomarker levels or changes in biomarker levels to disease-free and overall survival, risk of metastases, recurrence of disease and response to treatment. Patient, tumor, and surgical characteristics will be collected for all patients included in this study. Differences between characteristics will be statistically compared between patients. In addition, these characteristics will be analyzed as possible predictors of survival using univariate and multivariate analyses through the log-rank test and Cox-proportional hazard models. These characteristics will include:

1. Patient and tumor clinical demographics including age, gender, race, primary tumor location, location of metastases, tumor grade, and Carbohydrate Antigen 19-9 (CA19-9) levels pre- and post-chemotherapy and pre- and post-resection.
2. Chemotherapy variables including type, number of cycles, response, and timing (neoadjuvant and/or adjuvant).
3. Operative variables including type of resection, number of lymph nodes resected, vascular proximity of tumor, histopathology, and post-operative mortality and complications.
4. Long-term variables including presence and location of recurrence and status (alive/dead).
5. Levels of specific biomarkers at the various time points of blood draws.
6. Numbers of circulating tumor cells at each time point in addition to genetic mutations present in the cells.

For all patients, overall survival will be calculated as the time between the date of the first pre-treatment blood draw to the date of the last known follow-up visit or death, whichever comes first. For patients who undergo surgical resection the investigators will calculate time to disease recurrence as measured from the day of surgery for pancreatic resection to the date of documented disease recurrence or death, whichever comes first. Time to disease recurrence and overall survival will be described with Kaplan-Meier curves and will be compared by log-rank test. The 1- and 5-year overall survival rate will be estimated with its 95% confidence interval. Statistical significance will be set for a p-value at or below 0.05.

The primary risks to this patient will be associated with the blood collection, but are no more than minimal given collection with be by venipuncture and no more than 50 ml will be taken at a time. The general risks associated with venous blood draw include bruising or minor bleeding at the site of venous puncture, which would be managed with the application of pressure.

There will be no direct benefit to patients in this study, as results generated will not be used for management of these patients. This study has the potential to benefit future patients by identifying methods for early diagnosis of patients with pancreatic cancer and predicting overall survival, response to treatment, or risk of metastatic spread, and predicting recurrence of disease, all of which has the potential advantage of improving outcomes for this deadly disease.

There will be no compensation given to subjects in this study. Costs in this study will be related to the cost of blood draws during various scheduled clinic visits. These will be minimal.

All patients who agree to participate in this study will be required to sign a consent form for the collection, storage, and use of their blood.