Real-World Study Reveals Reasons for Pancreatic Cancer Non-Treatment

Nicole S. Evans, MD

A retrospective cohort study of a real-world pancreatic cancer population found that many patients with pancreatic ductal adenocarcinoma (PDAC) decline treatment for their disease because of personal preferences, despite the survival benefits associated with treatment, according to Tilak Sundaresan, MD, and Nicole S. Evans, MD.

Data from the real-world study were presented at the 2024 ASCO Annual Meeting and showed that among all adult patients with PDAC (n = 3852) diagnosed between 2010 and 2020 in the Kaiser Permanente Northern California (KPNC) Registry, 77.7% of those with nonmetastatic disease and 56.2% of those with metastatic disease received any cancer-directed treatment.1 Among patients 65 years of age or younger and nonmetastatic disease, 90.3% saw an oncologist, and 89.3% were offered PDAC treatment. Reasons for declining treatment in this age group included patient/family preference (80.8%), PDAC-related frailty (11.5%), and medical comorbidities/other malignancies (7.69%). Similar trends were observed among the population of patients older than 65 years of age.

“It seems that patients are making these nuanced decisions about whether they want to receive treatment, even in a scenario where they have nonmetastatic disease,” Sundaresan, a medical oncologist and chief of Oncology and Hematology at the San Francisco Medical Center at Kaiser Permanente San Francisco in California, said in an interview with OncLive®.

“From a [gastroenterologist’s] perspective, we’re putting patients through an extensive diagnostic workup, and [maybe] we should be having some of these conversations about the implications of the results [of their diagnosis] earlier,” Evans, who is a gastroenterologist at Kaiser Permanente Southern California in Downey, added.

In the interview, Sundaresan and Evans discussed the reasons necessitating research into real-world factors affecting non-treatment rates among patients with pancreatic cancer, how the design of their retrospective cohort study provided opportunities for gathering nuanced data about this population, and the findings they uncovered through this research.

OncLive: What was the rationale for this research?

Sundaresan: This all started because we were working with some prior colleagues of mine at Dana-Farber Cancer Institute [in Boston, Massachusetts, to evaluate] patients who were diagnosed with pancreatic cancer and had CT scans or some sort of scan done prior to their diagnosis to determine whether there were any changes in body composition that preceded diagnosis. We had some interesting results and found that there were changes in body composition that occurred many years prior to [these patients’] diagnoses of pancreatic cancer. That work has since been published.

However, in the course of that project, we identified a population of patients within the KPNC system and found that approximately one-third of patients diagnosed with pancreatic cancer never went on to receive cancer-directed therapy. That intrinsically was a little surprising to me; it was a higher proportion of patients than I had expected to see. As soon as we were done with that other project, we realized we needed to further investigate this group and better understand why these patients weren’t going on to receive treatment.

Evans: Pancreatic cancer is a leading cause of cancer-related mortality in the United States [US]. In and of itself, it is an important subject. With that, there’s a limited number of patients [with pancreatic cancer] who are even candidates for surgical treatment and curative treatment. Furthermore, a small proportion of patients with node-negative [disease upon] excision will end up still being cured 5 years later.

Some recent trials have shown that patients [with pancreatic cancer] who are offered treatment have improvement in their long-term survival. What about the patients who are electing not to get treatment but are offered it? What has been looked at in terms of that? It seems like not too much. We were curious about understanding why the patients chose not to get treatment when offered.

What were the methods and design of the study?

Sundaresan: Other studies have examined treatment rates for patients with a wide variety of different cancers. They’ve been fundamentally based within large databases, [such as the] Surveillance, Epidemiology, and End Results and Medicare databases, to see which patients receive treatments. [These studies have been conducted under] the framework of receiving guideline-concordant care.

We thought we had a novel opportunity within the KPNC system, because we have a huge population of patients and a large, well–clinically annotated database for these patients, but also the ability to look into the records for these patients and better understand what their interactions with the clinical system were like. Did they see oncologists? How did those conversations with oncologists go? We felt we could add more nuance. That was the driving imagination behind this project.

We [observed] a retrospective cohort. We included all patients diagnosed with pancreatic cancer from 2010 to 2020 within the KPNC system. At KPNC, we take care of approximately 4.5 million patients in northern California and diagnose and treat approximately 1% of all the pancreatic cancers that are diagnosed in the US—so, [this is] a large group of patients. We evaluated all the patients who had histologically confirmed pancreatic adenocarcinoma. We subsequently did a multivariable logistic regression analysis to see what covariates were associated with the patients who did not receive treatment.

Evans: This was a retrospective cohort study. We used our integrative health care system through KPNC in conjunction with our KPNC cancer registry to gather all the data. In general, we used the data to look at trends and compare metastatic vs nonmetastatic disease and any treatment vs surgical and other [specific] treatments to get numbers.

However, what ended up being the more interesting part of our study was the chart review we did to understand patients’ rationales for declining treatment, specifically those with nonmetastatic disease who were under the age of 65 years. We broke that down into broad buckets. We evaluated patient/family preference as one option, poor prognosis and disease frailty [as another category], and other significant medical comorbidities or other primary malignancies. We found that those were, in general, the large buckets of reasons why patients elected not to undergo treatment.

What were the key findings from this study?

Sundaresan: The topline [data revealed] the proportion of patients who didn’t receive treatment. [We observed] that approximately one-third of all patients diagnosed with pancreatic cancer did not receive treatment. We were most interested in [answering]: Why didn’t those patients receive treatment, and how did their care journey [lead] to them not getting treatment?

We didn’t understand. Were these patients not receiving treatment because they didn’t see an oncologist? [Were] they diagnosed in the emergency department and [their health care providers] felt they were too sick, and so they just never made it further than that and were going straight to hospice or something? Or were they having a conversation with their primary care doctor and having a sense of nihilism about the diagnosis of pancreatic cancer and didn’t want to go any further and never see an oncologist? We wanted to understand. How far along in the [treatment] process did they go?

We were surprised to see that most of these patients chose to get a biopsy. A lot of people have a mass in their pancreas and never get histologic confirmation of pancreatic cancer; they just stop right there and choose to never get a biopsy. That’s a different population; those are oftentimes people who are far more sick, [such as those] diagnosed in the Intensive Care Unit. However, in this group of patients, we found that firstly, the vast majority, 90.3%, went on to see an oncologist, so they had a consultation with an expert to talk about the diagnosis and think about treatment options.

We wanted to understand: How did this conversation with the oncologist go? Did the oncologist meet with them and say they were too sick, and they didn’t think the patient would do well with chemotherapy or whatever treatment and that was the rationale for this group of patients not getting treatment? Surprisingly to us, that was not the case. Among this group of patients who eventually did not receive treatment, not only did 90.3% of them see an oncologist, but most of them, after seeing an oncologist, received a recommendation from the oncologist to proceed with treatment: either palliative systemic therapy if they had metastatic cancer, or potentially further exploration of surgery if they had nonmetastatic pancreatic cancer. Most patients, after seeing an oncologist, received a recommendation from their oncologist to go on to receive cancer-directed therapy.

I was not expecting to find that [the main reason for] this group of patients not receiving treatment for pancreatic cancer was a patient-directed phenomenon. These were patients choosing not to receive treatment. It made us think more about understanding this decision-making process because that’s an interesting finding. It goes a bit against the narrative that I might have expected leading into this, which is that there’s this clinical nihilism about managing patients with pancreatic cancer, and much of that might be driven from the provider side. That was interesting to us. It’s a big part of what we’re thinking about in terms of future research possibilities and where we can go from here.

Evans: In general, in all patients, we found that increasing age correlated with decreased treatment rates, as one would expect. Patients with metastatic disease also had increased comorbidities that correlated with decreased treatment [rates]. With Kaiser Permanente being an integrated health care system, we found that race and gender were not associated with changes in treatment rates. That was an important finding. Irrespective of age, a significant proportion of patients chose to forego treatment despite the known survival benefits. Specifically, the younger patients who did not undergo treatment did so primarily due to patient/family preference.

Reference

1. Sundaresan TK, Evans NS, Gupta S, et al. Treatment rates and reasons for non-treatment in a real-world cohort of patients with pancreatic cancer. J Clin Oncol. 2024;42(suppl 16):e13554. doi:10.1200/JCO.2024.42.16_suppl.e13554