Participants in the PanCAN Patient Registry were patients with pancreatic cancer or their caregivers, identified through PanCAN’s Patient Services Help Line. Participation was voluntary and required informed consent for the use of their data in research. Patients and caregivers could independently enroll in the PanCAN Registry through the PanCAN website. Upon creating a profile and signing an online informed consent form, participants completed surveys that documented their experiences with pancreatic cancer. Participants completed surveys providing detailed information on diagnosis, symptoms, treatments, complementary medicine regimens, health care decisions, and more.

There have been 2 versions of PanCAN’s Patient Registry. The results reported in this analysis are based on PanCAN Registry version 2, which was open for enrollment from December 2020 through January 2024. The data collected were facilitated by an online data vendor platform called LunaDNA, which housed the PanCAN Registry survey questions for participants to access. The change to version 2 was due to a transition in the vendor managing the PanCAN Registry database technology. Both registry version 1 and version 2 received institutional review board (IRB) approval through Genetic Alliance, and PanCAN updates the IRB annually to maintain registry study protocol compliance. Although PanCAN Registry version 1 and version 2 used different technology platforms, both functioned similarly as patient-facing databases and adhered to PEER requirements determined by Genetic Alliance.

We provide a Checklist for Reporting Results of Internet E-Surveys (CHERRIES) (Checklist 1) that further describes the platform, the development and testing of questions, marketing, data protection, and more [10].

PanCAN staff worked with LunaDNA, the platform vendor, to transpose the surveys into the proper platform formatting, including branching logic and data extraction. The surveys used in version 2 of the PanCAN Registry were previously developed in version 1 of PanCAN’s Patient Registry and used with occasional changes or updates. These pancreatic cancer-specific surveys were developed and reviewed by experts in the domain and patients affected by pancreatic cancer. The experts included PanCAN staff, oncologists, gastroenterologists, scientists, a dietitian, and a radiation oncologist. The General Assessment survey, previously the Health Assessment survey in PanCAN Registry version 1, included questions derived from the Patient-Reported Outcomes Measurement Information System (PROMIS)-29 validated survey [11,12].

Data were extracted from the online LunaDNA-hosted registry (PanCAN Patient Registry version 2). Bivariate analysis was conducted using the chi-square test for categorical variables.Statistical significance was defined as a P value of <.05, with P values between .05 and .1 considered marginally significant, and P values >.1 considered insignificant. Due to the relatively small sample size of this study, a significance level of .1 was used to draw conclusions. While a P value of <.05 is a conventional threshold in biomedical research, the use of a .1 threshold is sometimes used in social science research, where increasing sample sizes is not always feasible. The Social Science Statistics calculator includes significance level options of .01, .05, and .10 [13]. While a P value of <.05 is a conventional threshold in biomedical research, in this context, the 10-fold difference between a P value of .06 and .6 is considered meaningful, and we optimized the significance level for this study as per Mudge et al [14].

Efforts to increase the sample size were not possible due to the unexpected closure of the LunaDNA platform, limiting further recruitment. In addition, publication of results from the PanCAN Registry version 2 is necessary to fulfill patient consent requirements and facilitate further analysis of the data. All user response data collected was deduplicated by identifying unique subject IDs within the deidentified data set. This information was organized in tables to display responses such as demographics, interest in joining the registry, sex, age, and more. All data were manually reviewed and validated by PanCAN staff.

Participants could complete up to 7 unique surveys on the PanCAN Registry website, totaling approximately 175 questions if all surveys were completed. Participants were required to complete the Basics survey before accessing additional surveys. The Basics survey gathered information about the person filling out the survey, the patient’s diagnosis and experiences with pancreatic cancer, and high-level information about symptoms, treatments, and reasons for participation. For this study, we defined users as PanCAN Registry participants who had completed at least the Basics survey.

The technology, user interface, regulatory requirements, and IRB compliance for the PanCAN Registry platform technology have been previously described [3]. The adherence to IRB requirements for the PanCAN Registry platform technology and the collaboration with LunaDNA to ensure the confidentiality and integrity of the data were described in Gupta et al [3]. All patients that joined the platform to participate in the study had the opportunity to remove their data if they chose. This is why LunaDNA reinforced the use of a sandbox workbench when the protocol was active, and participants were enrolling. However, it was explained to participants that they were not able to remove data that was part of a downloaded research set for publication purposes.

The Patient Registry received approval for Protocol PCAN001 from the Genetic Alliance IRB on January 19, 2024, as part of its annual review process. Since its launch in 2015, the registry consistently maintained compliance with IRB requirements as determined by the Genetic Alliance.

As described in the informed consent, data privacy and security were central to the registry’s operations. In this agreement with LunaDNA, genomic data (ie, data about an individual’s genes or DNA) and medical or health data (eg, medications, allergies, surveys, health records, and information collected by integrated apps and devices) were referred to as Shared Data. To protect participant privacy, Shared Data were separated from Personal Data, a process referred to as deidentification. Once deidentified, Shared Data were aggregated with data from other participants to create a searchable database designed to support research and discovery while protecting individual privacy.

As outlined in the participation and enrollment section, individuals who wished to join the registry had to create a profile and electronically sign an informed consent form before completing surveys about their pancreatic cancer experiences. Those who chose not to sign the informed consent were not eligible to participate.

Participants may revoke their consent or request deletion of their account at any time, in which case their data will be permanently removed or purged from the database. However, any research already conducted or published using the participant’s data before revocation of consent or data deletion will remain unaffected. Participants did not receive compensation for participation in the patient registry.

The demographics of the patient population who participated in version 2 of the PanCAN Patient Registry are shown in Table 1. During the time period analyzed, 272 individuals filled out the basics survey in the LunaDNA-based PanCAN Patient Registry. Of the 191 participants who indicated their age, 1 participant (0.5%) was 11‐15 years old, 13 (6.8%) were aged 25‐44 years, 89 (46.6%) were aged 45‐64 years, and 88 (46.1%) were aged 65 years and above. For the purpose of the analyses described below, we stratified patients as under 65 years (53.9%) or 65 years and older (46.1%).

These 191 participants were evenly distributed by sex, with equal numbers identified as female and male at birth. The population had minimal racial and ethnic diversity, with 191 respondents providing 210 answers (multiple options were allowed). The majority (171/210, 81.4%) identified as White, and 11/210 (5.2%) of participants identified as being of Hispanic, Latino, or Spanish origin, and 7/210 (3.3%) identified as Black or African American.

Of the 272 total participants, 102 (37.5%) were initially diagnosed with metastatic pancreatic cancer, 38 (14%) with locally advanced disease, 35 (12.9%) borderline resectable, and 75 (27.6%) had resectable pancreatic cancer at diagnosis. The remaining 22/272 (8.1%) of respondents were unsure of their stage of disease at diagnosis. It is worth noting that the average distribution of disease stage at diagnosis is 51% metastatic and 14% localized [15], so the patient population in this study was skewed toward earlier stage disease compared with the overall patient population with pancreatic cancer.

Participants were also asked to indicate their reasons for joining the PanCAN Patient Registry. Multiple answers could be selected, and all 272 participants responded to this question. The majority (255, 93.8%) of responses indicated that the participant joined the registry “to provide information for researchers and other patients,” showing a deep sense of altruism. The next most common answer (231, 84.9%) was “to learn more about pancreatic cancer.” A majority (162, 59.6%) of responses indicated that the participant felt the registry would help them “to share information with friends, family, or a doctor.”

Pancreatic cancer and its treatments are known to cause significant pain, typically of the abdominal area and lower back. Participants in the PanCAN Patient Registry were asked several questions pertaining to their experience with pain within the 7 days before their responding to the survey. A total of 7 questions addressed the presence and intensity of pain as well as its interference with day-to-day activities (Supplementary table in Multimedia Appendix 1). For the purpose of this analysis, we stratified the responses to a yes or no response in regard to the participants experiencing pain over the week before filling out the survey (Table 2).

Using this methodology, we found that out of 207 respondents, 186 (89.9%) reported pain within the previous 7 days. There was no difference based on sex; approximately 90% of both male and female respondents reported pain.

There was, however, a marginally statistically significant (P<.1) difference found in the reporting of pain by age groups, with pain more frequently reported by younger patients. In those aged younger than 65 years, 86/92 (93.5%) reported experiencing some pain over the previous 7 days. A lower percentage (66/78, 84.6%) of individuals aged 65 years and above reported experiencing pain (P=.06).

Depression is also frequently experienced by people with pancreatic cancer, as shown in Table 2. For this standalone survey, participants were asked whether they were feeling or had felt “depressed at any time throughout the course of the disease.” There were 103 respondents to this question, and 64 (62.1%) indicated that they were feeling or had felt depressed, 35 (34%) indicated no depression, and 4 (3.9%) were unsure. There were no statistically significant differences in the responses to feeling or had felt depression by sex or by age.

Finally, we were interested in determining the concurrence of pain and depression experienced by individuals who filled out the PanCAN Patient Registry. We hypothesized that those experiencing pain would be more likely to indicate feelings of depression. Indeed, a majority (53/80, 66.3%) of individuals who indicated that they felt pain within 7 days of filling out the survey also said they were experiencing or had experienced feelings of depression. Among individuals who reported no pain, 5/13 (38.5%) answered that they experienced depression. This difference approached statistical significance, with a P value of .07.