The SPP study is a prospective population-based case-control study. We include incident cases of pancreatic cancer in the Stockholm county region during 2008 to 2014. Controls are randomly selected from general population in the Stockholm region, and individually matched to cases by age and sex.
Cases selection. The criteria for cases selection including those are age less than 85 years, born in Sweden and presently living in the county of Stockholm. The cases of pancreatic cancer are first identified within the framework of the clinical healthcare system of the Stockholm County. Through a network, consisting of all surgical clinics and the only oncological clinic in the Stockholm county area, we were able to include a vast majority of cases diagnosed with radiological methods, he or she were asked to participate in the study. After informed consent is obtained from the patient, the study coordinator at the Karolinska University Hospital at Huddinge was notified. The coordinator will inform Statistics Sweden interviewer to interview the case patient. Further, after radiological diagnosis of pancreatic cancer, we try to retrieve a tumor biopsy for every case. In total around 30% of the cases are surgically resected and of the remaining 70% of the patients treated with a palliative intention and among which more than 80% are biopsied using fine needle aspiration (the remaining 20% not biopsied for practical reasons such as clinical deterioration), thus resulting in a total proportion of patients in which tumor tissue is retained of close to 90%.
Control selection. The control persons (non-cancer patients) are randomly identified via the Statistics Sweden. For each case, a gender- and age-matched (same birth year) control person will be identified randomly from the general population of the Stockholm County.
Data collection. Through professional interviewers from the Statistics Sweden, interviews were conducted concerning environmental factors with both cases and control persons. As soon as the study coordinator identified a case of pancreatic cancer and the patient has accepted to participate, Statistics Sweden is contacted about this patient's identify and location. Professional interviewers, blinded to the research hypotheses of the study, will then contact and go to the patient at home, at the hospital or wherever the patient is and perform the interview which is calculated to take around 1 hour and 15 minutes. The questionnaire contains items concerning demographics, socioeconomic factors, physical activity, height and weight history, tobacco and snuff use, dental health, earlier and concomitant diseases and medication history. A structured food frequency questionnaire, a modified version of a previously evaluated questionnaire, including 63 food and beverage items of interest, are used to evaluate dietary habits 20 years prior to interview.
Once a case has been identified, a set of up to 5 controls (same sex and birth year) from Stockholm general population will be selected by the Statistics Sweden. The control persons in the list are contacted by order, until a control person agrees to participate in our study. The interviewers will perform the interview in exactly the same way for the cases.
For each case and control, if they agree, 20 ml of blood are collected. The samples are separated into serum and cells (red and white together) and aliquot into 4 separate serum (for subsequent protein analysis) samples and 2 cell samples after Ficoll gradient (for subsequent DNA and RNA analysis) and immediately frozen to -80°C. Control persons that decline leaving a blood sample are instead offered to leave a saliva sample, from which DNA can be extracted and analyzed.
