Pre-existing psychological disorders, diabetes, and pancreatic cancer: A population-based study of 38,952 Finns

1Introduction

Pancreatic cancer is one of the most fatal cancers with the worst prognosis, largely because it is an aggressive form of cancer that has few symptoms until it becomes advanced [1] . Potential risk factors for pancreatic cancer include tobacco smoking, alcohol consumption, occupational exposures, obesity, diabetes, chronic infections, and certain genetic factors 234567 . However, many possible risk factors – such as pre-existing depression and anxiety – are still poorly understood 8910 . It has long been known that pancreatic cancer patients have high rates of depression and other psychological disorders 111213 . A review reported a prevalence of depression ranging from 33% to 50% in patients with pancreatic cancer, which was higher than the levels found in patients with other forms of cancer such as lung, breast, and oropharyngeal cancers [14] . This observation was confirmed in another review, which also noted the frequent coexistence of depression and anxiety in pancreatic cancer patients [8] . Depression has also been shown to be a precursor of pancreatic cancer [131516] . A large-scale study based on health claims of inpatients and outpatients in the US found that the risk of pancreatic cancer doubled in men with mental disorders, and that these disorders were, on average, diagnosed 1.5 years before the pancreatic cancer diagnosis; however, this was not observed in women [17] .

Nonetheless, researchers have called into question the direction of the relationship: whether depression and anxiety are precursors to pancreatic cancer, or whether these conditions are psychological consequences of being diagnosed with pancreatic cancer [813] . Patients’ psychological distress may start to increase when they are undergoing medical examinations and are suspected of suffering from pancreatic cancer. Thus, studies that examine the associations between the timing of the onset of depression and anxiety and pancreatic cancer risk are both necessary and important. In addition, there is evidence that the association between having diabetes and pancreatic cancer risk varies depending on when the patient was diagnosed with diabetes [1819] . The risk of pancreatic cancer associated with recent-onset diabetes (e.g., 2–3 years prior to the diagnosis of pancreatic cancer) has been found to be greater than that associated with long-term diabetes [1820] . A meta-analysis showed that pancreatic cancer risk decreased as the duration of diabetes increased [21] . A similar gradient may also exist between the timing of the onset of depression and anxiety and pancreatic cancer risk.

The risk factors for pancreatic cancer may also affect the clinical characteristics of pancreatic cancer and its treatments and outcomes. Cancer patients with pre-existing depression, including pancreatic cancer patients, have been found to have worse survival than those without depression 222324 . The relationships between pre-existing mental disorders and the cancer stage and treatments remain unclear [25] . Similarly, although an association between diabetes and pancreatic cancer survival has been previously reported [2627] , no consensus has been reached on whether the timing of the onset of diabetes influences survival [2829] . Therefore, there have been calls for further research on the relationship between diabetes and the prognosis of pancreatic cancer [30] .

The present study, using longitudinal Finnish registry data from 1995 to 2015, investigated (1) how pre-existing depression, anxiety, and diabetes with different durations affect the risk of pancreatic cancer, and (2) how having these pre-existing conditions influences the stage of pancreatic cancer, the treatment modalities, and the survival of pancreatic cancer patients.