GRAPHICAL ABSTRACT

Highlights

• PPDM had a higher pancreatic cancer risk than T2D or pancreatitis alone.

• Among PPDM patients, shorter DM duration or insulin use conferred an elevated risk.

• These findings highlight the need for close monitoring in high-risk PPDM groups.

INTRODUCTION

Pancreatic cancer was ranked twelfth in terms of incidence (495,773 new cases), but was the seventh leading cause of cancer mortality (466,003 deaths) worldwide in 2020 [1]. In South Korea, cancer has emerged as the leading cause of death among patients with diabetes, accounting for about 30% of all deaths [2], and in the overall population, pancreatic cancer was the eighth most commonly diagnosed cancer (8,414 new cases), with the fifth highest age-standardized cancer mortality rate (5.7 per 100,000) in 2020 [3]. The Surveillance, Epidemiology, and End Results (SEER) 5-year relative survival rates for pancreatic cancer during 2013–2019 were 44.3%, 16.2%, and 3.2% for localized, regional, and distant stages, respectively, with the survival rates declining sharply as the staging progressed [4]. In most cases, pancreatic cancer is diagnosed at the distant stage, while only some cases are diagnosed at the local stage (13.0%) [3,4]. One of the reasons pancreatic cancer has one of the highest mortality rates is that, it is mostly diagnosed in advanced stages owing to the lack of effective screening strategies.

Although diabetes [5-7] and pancreatitis [8-13] are known risk factors for pancreatic cancer, current guidelines do not recommend aggressive screening in patients with diabetes or pancreatitis because pancreatic cancer screening is specifically recommended for high-risk populations, including individuals with a >5% lifetime risk or a relative risk increased by more than 5-fold compared to the general population [14,15]. Previous studies on the risk factors for pancreatic cancer have attempted to simultaneously analyze diabetes and pancreatitis [16-18]. Population-based cohort studies in Taiwan found that patients with both diabetes and chronic pancreatitis had a significantly increased risk of pancreatic cancer compared to those with neither disease [16,18]. Another population-based cohort study in South Korea found that patients with acute pancreatitis had a higher risk of pancreatic cancer in patients with type 2 diabetes mellitus (T2DM) [19]. In addition, a study based on the nationwide cancer registry in New Zealand reported that post-pancreatitis diabetes mellitus (PPDM) was associated with a significantly higher risk of pancreatic cancer than T2DM alone [17]. Although previous studies provide important insights, these studies either did not consider the sequential occurrence of diabetes and pancreatitis or lacked comparisons with the general population.

This nationwide, population-based study pursued three main objectives: (1) to examine the combined effect of pancreatitis and diabetes on the risk of pancreatic cancer relative to their absence; (2) to determine if the sequence of onset between diabetes and pancreatitis influences the risk of pancreatic cancer; and (3) to pinpoint high-risk subgroups for pancreatic cancer, for whom screening might be advised.

METHODS

Data sources

The data used in the current study were extracted from the National Health Information Database (NHID) of the Korean National Health Insurance Service (NHIS) for the period between 2002 and 2021 [20]. The Korean NHID contains nationwide claims and health examination data [21]. The NHIS, as the single insurer, pays health care costs based on the billing records of health care providers [20]. The NHIS covers almost the entire population of South Korea, with over 50 million individuals as of 2020 [22]. The NHID provides researchers with a platform for using customized retrospective cohort data [23]; moreover, it contains information on income-based insurance contributions, demographic variables, and date of death, as well as information regarding records on inpatient and outpatient usage (diagnosis codes, hospital admission, and special reimbursement codes) and medical treatment (prescriptions and procedures) from January 2002 to December 2021 [23]. This study adhered to the ethical standards set forth in the Declaration of Helsinki. The study protocol was approved by the Institutional Review Board (IRB) of Yonsei University College of Medicine (No. 4-2021-1489). Given the use of de-identified data from a national database, the IRB granted a waiver of informed consent.

Study population

Data were collected from 2002 to 2021, but because of a 5-year washout period, we enrolled patients newly diagnosed with pancreatic cancer from 2007 to 2021 as the case group and matched patients who did not develop pancreatic cancer as the control group. The following exclusion criteria were applied: (1) diagnostic code of C254 (malignant neoplasm of endocrine pancreas); (2) history of any type of cancer (International Classification of Diseases, 10th Revision [ICD-10]: C00-97) in the 3 years prior to the date of pancreatic cancer diagnosis; (3) missing data (body mass index [BMI], alcohol consumption, and smoking); (4) type 1 diabetes mellitus (T1DM) (ICD-10: E10) without pancreatitis or T1DM prior to pancreatitis; and (5) concurrent diagnosis of diabetes and pancreatitis within 3 months before the index date or hospital visit in 2002 for both chronic pancreatitis and diabetes. Because the database did not contain claims data prior to 2002, it was not possible to determine which disease preceded the other.

For the selection of case observations, we first selected patients with newly diagnosed pancreatic cancer from 2007 to 2021 and excluded those who met the exclusion criteria. Then, we matched 60,716 patients with pancreatic cancer to 6,071,491 observations without pancreatic cancer (control) by birth, sex, and length of follow-up at a ratio of 1:100. After incidence density sampling, we excluded observations in the control group who met the exclusion criteria and conducted simple random sampling at a ratio of 1:9. Finally, the control group included 546,444 observations (Fig. 1).

Clinical variables

Diagnosis of diabetes and pancreatitis was only considered if they occurred before the index date. We defined the index date as 1 year prior to the date of diagnosis of pancreatic cancer because pancreatitis and hyperglycemia can be symptoms of pancreatic cancer development [24,25]. Although the index date was set as 1 year prior to the date of diagnosis of pancreatic cancer for the main analyses, in the analysis that further divided the groups according to diabetes mellitus (DM) duration and pancreatitis duration, the index date was defined as the date of diagnosis of pancreatic cancer to include new-onset disease.

Pancreatic cancer was defined when ICD-10 code C25 and special reimbursement code V193 were recorded simultaneously. The NHID started using special reimbursement codes (V193 codes) since September 2005 to reduce the copayment rate to 5% for cancers. In order to register a cancer patient under the V193 code, the diagnosis must be confirmed by a test (such as histologic, cytologic, biochemical, immunohistochemical, hematologic, imaging, or diagnostic surgery) based on the criteria established by the Ministry of Health and Welfare. Therefore, the use of ICD-10 C-codes and V193 codes allows for an accurate definition of cancer.

Patients were deemed to have diabetes if glucose-lowering medications were prescribed and the ICD-10 code for diabetes (ICD-10: E10-14) was present in inpatient or outpatient NHIS claims data [26]. Patients with T2DM were defined as patients without a T1DM diagnosis code (ICD-10: E10) before the index date in predefined patients with diabetes. Pancreatitis was confirmed when the ICD-10 code for pancreatitis was used, when acute pancreatitis (ICD-10: K85) was the principal diagnosis or first-to-fourth additional diagnosis on at least one inpatient NHIS claim, or when chronic pancreatitis was stated (ICD-10: K86.0 and K86.1) on at least one inpatient or more than two outpatient NHIS claims.

We divided the patients into five non-overlapping groups based on the grouping in a previous study [17,27,28]. The T2DM-alone (T2DM-only) group included the patients who were diagnosed with T2DM and had never been diagnosed with pancreatitis prior to the index date. The pancreatitis-only (PAN-only) group included patients who were diagnosed with pancreatitis and had never been diagnosed with diabetes prior to the index date. Patients with both diabetes and pancreatitis prior to the index date were divided into two groups based on the sequence of disease onset and type of diabetes. The pancreatitis after T2DM (T2DM-PAN) group included those who were diagnosed with pancreatitis at least 90 days after diagnosis of T2DM. The PPDM group included those who were diagnosed with diabetes (ICD-10: E10-14) at least 90 days after the first diagnosis of pancreatitis. Those who did not fall into any of the above four disease categories were categorized into the no diabetes and no pancreatitis (NDNP) group.

Smoking status was classified as current, past, or non-smokers based on participants’ responses to the health screening questionnaire. Alcohol consumption was estimated based on survey responses, calculated in grams per week, converted to standard drinks (14 g of alcohol per drink), and categorized as ‘heavy drinking’ if men consumed 15 or more drinks per week and women 8 or more [29].

The risk of pancreatic cancer for each subgroup was analyzed. The PPDM, T2DM-PAN, and T2DM-only groups were further stratified based on insulin use into continuous, temporary, and non-user categories. Continuous insulin users were defined as those who received a minimum of five insulin prescriptions within the 3 years preceding the index date (for hospitalizations, multiple prescriptions during a single admission were counted as one instance). Temporary insulin users were defined as those who had received at least one insulin prescription but did not fulfill the criteria for continuous use. Furthermore, we conducted a sensitivity analysis by redefining the continuous insulin user criteria as receiving insulin prescriptions either (1) at least three times or (2) at least twice within the 3 years prior to the index date.

Additional analyses for the duration of diabetes and pancreatitis were performed by only including diabetes and pancreatitis cases diagnosed after 2004 to include subjects for whom the exact duration of diabetes and pancreatitis could be accurately determined, and by using the index date as the date of pancreatic cancer diagnosis (without lag period) to include new-onset DM and new-onset pancreatitis cases with a duration of less than 1 year.

Sensitivity analysis

To minimize the potential impact of reverse causality, a 1-year lag period was applied when setting the index date in the main analysis. Additionally, we conducted a sensitivity analysis using various lag periods (0, 0.5, 1, 2, 3, and 5 years) to evaluate whether the outcomes were influenced by the length of the lag period. Additionally, we performed another sensitivity analysis by excluding individuals who had undergone pancreatectomy prior to the index date.

Statistical analysis

We first compared the baseline characteristics of case and control groups based on the diagnosis, using chi-square tests for categorical variables and t-tests for continuous variables. Baseline characteristics consisting of continuous variables were presented as mean±standard deviation, and categorical variables were presented as number (%).

We performed the analyses using a conditional logistic regression model to determine whether T2DM-only, PAN-only, T2DM-PAN, or PPDM status was associated with a higher risk of pancreatic cancer than that related to the NDNP status. Multiple logistic regression was performed using BMI, tobacco use, and alcohol consumption as variables. We used SAS version 9.3 (SAS Institute, Cary, NC, USA) for data analysis.

RESULTS

Baseline characteristics

As shown in Table 1, in the majority of the cases, pancreatic cancer was diagnosed in people aged >70 years, and men were more frequently diagnosed with pancreatic cancer than women (54.5% vs. 45.5%). The proportion of heavy drinkers (7.3% vs. 6.6%) and current smokers (22.1% vs. 16.8%) was higher in the pancreatic cancer group than in the control group. The difference in the mean BMIs of the case and control groups was statistically significant, but the mean BMIs were numerically similar at 24.1 and 24.0 (Table 1). In all groups except the NDNP group, the mean age was higher in the non-pancreatic cancer group, whereas the percentage of older adults was lower in the pancreatic cancer group compared to the control group (Supplementary Tables 1 and 2).

Association of DM and/or pancreatitis with pancreatic cancer risk

In the conditional logistic regression analysis, PPDM and T2DMPAN were significantly associated with pancreatic cancer, with higher odds ratios than those for the T2DM-only and PAN-only groups. In the simple logistic regression, the adjusted odds ratios (AORs) were significantly higher in the T2DM-only (AOR, 1.65; 95% confidence interval [CI], 1.61 to 1.68), PAN-only (AOR, 2.96; 95% CI, 2.80 to 3.13), T2DM-PAN (AOR, 4.96; 95% CI, 4.48 to 5.49), and PPDM (AOR, 4.71; 95% CI, 4.12 to 5.37) groups than in the NDNP group (Fig. 2, Supplementary Table 3). Both being a current smoker and having a heavy drinker status were significantly associated with the incidence of pancreatic cancer (Fig. 2).

We conducted analyses by alternately using the T2DM-only, PAN-only, and T2DM-PAN groups as the reference. Among the four groups, both PPDM and T2DM-PAN demonstrated a significantly elevated risk of developing pancreatic cancer compared to the PAN-only or T2DM-only groups (Table 2). However, in direct comparisons, no significant difference was observed between the PPDM and T2DM-PAN groups (AOR, 0.95; 95% CI, 0.81 to 1.12) (Table 2).

Pancreatic cancer risk in DM subgroups based on insulin use

Additional analysis of the PPDM, T2DM-PAN, and T2DMonly groups by insulin use showed that in all cases, the risk of pancreatic cancer tended to be higher in patients with diabetes using insulin continuously than in those not using insulin in all the subgroups. Specifically, compared to the NDNP group, the PPDM group, which included individuals who had used insulin, exhibited a 17-fold increase in pancreatic cancer risk (AOR, 16.72; 95% CI, 9.50 to 29.43), whereas, the T2DM-PAN group, which included individuals who had used insulin, demonstrated a 8-fold increase (AOR, 7.78; 95% CI, 5.73 to 10.58) (Table 3). When the threshold for defining continuous insulin use was adjusted from at least five prescriptions within 3 years before the index date to three or two prescriptions, the odds ratio for pancreatic cancer in the continuous insulin user group exhibited a modest decrease. Nevertheless, the risk of pancreatic cancer remained significantly elevated among continuous insulin user groups within PPDM, T2DM-PAN, and T2DM-only, highlighting a consistent trend across these subgroups (Supplementary Tables 4 and 5).

Analysis by duration of DM or pancreatitis

We subdivided the PPDM and T2DM-only groups according to the duration of diabetes and found that the shorter the duration of diabetes, the higher the risk of pancreatic cancer. In the T2DM-only group, new-onset T2DM (DM duration <365 days) was associated with a higher risk of pancreatic cancer than long-standing T2DM (DM duration >1,825 days) (AOR 9.10; 95% CI, 8.74 to 9.48 vs. AOR, 1.71; 95% CI, 1.66 to 1.76) (Table 4). Additionally, the PPDM group, with a diabetes duration of 3 years or less, had a more than 8-fold increased risk of pancreatic cancer compared to the NDNP group (Table 4). We conducted further analysis using PPDM or T2DM-only patients with a longer history of diabetes (DM duration >1,825 days) as the reference group and found that PPDM and T2DMonly patients with a shorter history of diabetes (duration less than 3 years) had a significantly higher association with pancreatic cancer risk (Supplementary Table 6). T2DM-PAN and PAN-only groups were further subdivided according to pancreatitis duration, and new-onset pancreatitis was associated with a higher risk of pancreatic cancer in both groups (Supplementary Table 7).

Sensitivity analysis

The risk of pancreatic cancer was relatively higher in the PPDM, T2DM-PAN, PAN-only, and T2DM-only groups when the lag period was shorter than 1 year (no lag period or 180-day lag period) compared to when the lag period was longer than 1 year (lag periods of 1, 2, 3, or 5 years). Whereas, the PPDM, T2DM-PAN, PAN-only, and T2DM-only groups maintained high ORs for pancreatic cancer when the lag period was increased to more than 1 year (Supplementary Table 8).

To further validate our findings, we conducted a sensitivity analysis by excluding individuals who had undergone pancreatectomy prior to the index date (311 individuals in the cancer group and 96 in the control group). The results of this analysis were consistent with those identified when pancreatectomy cases were included, indicating that the exclusion of these individuals did not significantly alter the observed associations (Supplementary Table 9).

DISCUSSION

The main finding of our study was that the coexistence of diabetes and pancreatitis (PPDM and T2DM-PAN groups) increased the risk of pancreatic cancer more than the existence of T2DM-only or PAN-only did. In this study, compared to the NDNP group, both the T2DM-only and PAN-only groups showed a significantly increased risk of pancreatic cancer. Additionally, both the T2DM-PAN and PPDM groups had a significantly higher risk of pancreatic cancer than did the T2DMonly or PAN-only groups. These results suggest that T2DM and pancreatitis independently serve as risk factors for pancreatic cancer, regardless of the order in which they occur, and that the concurrent presence of both diseases is associated with a higher risk of pancreatic cancer compared to having only one of the diseases.

Considering that pancreatic cancer screening is recommended for high-risk populations, defined as individuals with a 5-fold increased relative risk [14], we further analyzed the PPDM, T2DM-PAN, and T2DM-only groups with respect to insulin usage and duration of diabetes to pinpoint subgroups that were associated with more than a 5-fold increase in pancreatic cancer risk compared to the general population. Firstly, we found that insulin usage before the index date was associated with a significantly higher risk of pancreatic cancer compared to lack of insulin usage. PPDM, when accompanied by continuous insulin usage, presented a 17-fold increased risk of pancreatic cancer, and T2DM-PAN with continuous insulin use demonstrated a 8-fold increased risk. The reason behind this finding could be that the use of insulin indicates difficulty in controlling blood sugar, which may reflect hyperglycemia caused by β-cell failure in the progression of diabetes [30,31] or reverse causality of pancreatic cancer [32]. Particularly, PPDM with continuous insulin use was associated with an approximately 17-fold increased risk of pancreatic cancer compared with NDNP (AOR, 9.84; 95% CI, 6.77 to 14.29), suggesting the need for careful surveillance for pancreatic cancer in patients with PPDM on insulin. Secondly, we found that the PPDM group with new-onset diabetes exhibited significantly stronger associations with pancreatic cancer than did the NDNP group. The T2DM-only group had a 9-fold increased risk of pancreatic cancer compared to the NDNP group up to only 1 year after diabetes onset and about 2-fold thereafter, while the PPDM group was associated with over 8-fold increased risk of pancreatic cancer compared to that of NDNP for a duration of up to 3 years after diabetes onset. These findings are consistent with previously published research, which reported that individuals with new-onset diabetes have a substantially higher risk of developing pancreatic cancer compared to those with long-standing diabetes. The concordance of our results with previous research highlights the critical need to consider the temporal relationship between diabetes onset and pancreatic cancer risk in clinical practice [33]. Furthermore, as supported by Sharma et al. [24], hyperglycemia and new-onset diabetes may precede pancreatic ductal adenocarcinoma diagnosis by several years, suggesting diabetes might serve as an early marker of pancreatic cancer.

Although the evidence presented previously is limited to acute pancreatitis, diabetes that develops after pancreatitis is likely multifactorial, with a variety of potential contributing factors in individual patients, including (1) loss of islet cell mass due to acute pancreatitis; (2) autoimmunity due to acute pancreatitis; (3) common risk factors for acute pancreatitis and diabetes; (4) local and systemic inflammatory responses; and (5) alterations in the insulin-incretin axis, or a combination of these factors [34]. Furthermore, a previous systemic meta-analysis showed that severe acute pancreatitis was associated with increased incidence of DM [35]. Although our study was not limited to acute pancreatitis, this suggests that new-onset diabetes after pancreatitis may have been associated with a higher risk of pancreatic cancer because new-onset diabetes after pancreatitis is itself a reflection of the severity of the pancreatitis.

In this study, no significant differences were observed when directly comparing the T2DM-PAN and PPDM groups (AOR, 0.95; 95% CI, 0.81 to 1.12), and the T2DM-PAN group had a significantly higher risk of pancreatic cancer than did the PAN-only group (AOR, 1.67; 95% CI, 1.49 to 1.88). The order of each disease’s onset did not significantly contribute to the risk of pancreatic cancer in patients with coexisting diabetes and pancreatitis. Importantly, our main finding demonstrated that the coexistence of these two conditions substantially elevated the risk of pancreatic cancer. However, these results were inconsistent with those of a previous study in which the PANonly and T2DM-PAN groups had similar risks of pancreatic cancer, and the PPDM group had a significantly higher pancreatic cancer risk than did the T2DM-PAN group [17]. This discrepancy may be ascribed to two principal factors. Firstly, the variation in the study populations is noteworthy. The prior study utilized data from the New Zealand Cancer Registry (NZCR), wherein, the patients not developing pancreatic cancer were subsequently diagnosed with other cancers. Moreover, the NZCR study encompassed a comparatively older demographic than did our study, with the mean age at diagnosis being 68.9 years for PPDM. Furthermore, Asian patients accounted for less than 10% of the study population. The etiology of pancreatitis (e.g., alcohol and tobacco use) and clinical outcomes (such as morphologic abnormalities and endocrine insufficiency) may vary by ethnicity [36]. Therefore, it can be hypothesized that the progression from pancreatitis to pancreatic cancer may also vary between different ethnic groups. Moreover, the incidence of pancreatic cancer also differs according to ethnicity and geographic location, likely due to differences in lifestyle, living standards, socioeconomic factors, and genetic predispositions [37]. Thus, the differences observed between our study and the New Zealand study could be attributed to interethnic variations in the development and progression of pancreatitis, as well as in the incidence of pancreatic cancer. Secondly, the distinction in study design played a significant role. T2DM-PAN and PPDM were delineated based on the sequence of disease onset, signifying an immortal time between the diagnosis of each condition. To circumvent the issue of immortal time bias in disease classification, we used a nested case-control design [38].

In our study, the lag period of the main outcome was set to 1 year to eliminate reverse causality, and the lag period was also set to 0, 0.5, 2, 3, and 5 years in the sensitivity analysis. Because high blood glucose can be a symptom of pancreatic cancer, consideration of reverse causality is necessary when evaluating diabetes as a risk factor in pancreatic cancer research. Notably, 74% to 88% of patients are diagnosed with diabetes within 24 months following their diagnosis of pancreatic ductal adenocarcinoma [39]. High serum glucose level may be one of the symptoms of pancreatic cancer, and the following explanatory mechanisms have been proposed: (1) diabetogenic factors secreted by the pancreatic tumor may reduce insulin secretion and sensitivity [40] and (2) β-cell dedifferentiation by pancreatic cancer may cause dysfunction of β-cells, leading to hyperglycemia [41]. Further, it is not known how many years it takes for symptoms to appear after the development of pancreatic cancer [42,43]. Hence, when diabetes is diagnosed before pancreatic cancer, it is difficult to determine exactly whether diabetes is a cause or a symptom of pancreatic cancer. In previous studies which have analyzed the relationship between diabetes or pancreatitis and pancreatic cancer, various lag periods were applied to exclude reverse causality [5,44]. In the sensitivity analysis of this study, the T2DM-only, PAN-only, T2DM-PAN, and PPDM groups exhibited a significantly increased risk of pancreatic cancer compared to the NDNP group at all lag periods. Importantly, the highest OR was observed in analyses conducted without a lag period conditions. However, even in sensitivity analyses with a lag period longer than 1 year, all groups were associated with an increased risk of pancreatic cancer compared to that of the NDNP group, suggesting that diabetes and pancreatitis not only have a reverse causality with pancreatic cancer but are also risk factors for pancreatic cancer.

A key strengths of this study is the higher number of pancreatic cancer cases compared to that of other studies, attributed to data derived from a comprehensive registry of the entire Korean population, a significant aspect considering the rarity of this disease. Furthermore, initiation of a special reimbursement program in September 2005 by the Korean NHIS, aimed at reducing the medical costs for patients with cancer, enhanced the study’s validity. As outlined in the methods section, patients were required to meet specific diagnostic criteria to qualify for special medical expense coverage. These stringent criteria ensured that the accuracy of pancreatic cancer diagnoses remained high, even when relying solely on insurance claims data. Additionally, utilizing the NDNP group as a reference enabled the evaluation of pancreatic cancer risk in the T2DM-only, PAN-only, T2DM-PAN, and PPDM groups compared to that of the general population. Importantly, the collection of data from medical examinations provided valuable insights into factors like smoking history, alcohol consumption, and BMI.

This study has some limitations that warrant discussion. Firstly, the histological type and stage of pancreatic cancer were not included in the NHID; therefore, these data could not be considered in our study. Secondly, discrepancies between medical diagnoses and diagnoses using operational definitions in claims data may have reduced the accuracy of the analysis. Therefore, we tried to improve the accuracy of the operational definitions of disease by using information such as hospitalization status, number of outpatient visits, medication use, and special reimbursement codes to define diseases [45]. Thirdly, our study lacks data on glycosylated hemoglobin (HbA1c) levels, which restricts our ability to assess the severity of diabetes and its potential impact on the development of pancreatic cancer. Future studies should incorporate HbA1c measurements to better evaluate this relationship. Lastly, due to the characteristics of the NHIS-NHID dataset, data prior to 2002 were unavailable, and therefore only patients who were newly diagnosed with diabetes from 2003 thereafter, for whom the duration of diabetes could be clearly established, were included in the analysis. Consequently, individuals with a diabetes duration greater than 20 years were excluded from this study. It is possible that if patients with longer diabetes durations had been included, they might have demonstrated a higher risk compared to those with newly onset diabetes; however, this could not be assessed due to the limitations of the dataset.

In conclusion, we demonstrated that the T2DM-PAN and PPDM groups had a higher risk of pancreatic cancer than did the T2DM-only and PAN-only groups. Importantly, even within the PPDM groups, a diagnosis of diabetes within 3 years or insulin use was associated with a significantly increased risk of pancreatic cancer, suggesting the need for aggressive pancreatic cancer screening for these patients. Because PPDM accounts for the second highest proportion of diabetes after T2DM [46], close follow-up is needed for the high-risk subgroup of patients with PPDM for pancreatic cancer development.

SUPPLEMENTARY MATERIALS

NOTES

CONFLICTS OF INTEREST

No potential conflict of interest relevant to this article was reported.

AUTHOR CONTRIBUTIONS

Conception or design: all authors.

Acquisition, analysis, or interpretation of data: all authors.

Drafting the work or revising: Y.K., E.S.K.

Final approval of the manuscript: all authors.

FUNDING

This study was supported by Big Data Research Funds from the Korean Society of Endocrinology in 2022, the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (No. RS-2023-00209104), a grant of the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI) funded by the Ministry of Health & Welfare, Republic of Korea (grant number: HI23C0923), and the Korean Endocrine Society for Convergence Research Award 2022.

ACKNOWLEDGMENTS

None

Fig. 1.

Flowchart of the study inclusion and exclusion criteria. T1DM, type 1 diabetes mellitus; DM, diabetes mellitus.

Fig. 2.

Crude odds ratio (OR) (95% confidence interval [CI]) and adjusted ORs for the association between pancreatic cancer and categorized disease groups. NDNP, no diabetes and no pancreatitis; PPDM, post-pancreatitis diabetes mellitus; T2DM-PAN, type 2 diabetes mellitus followed by pancreatitis; PAN-only, pancreatitis-only; T2DM-only, type 2 diabetes mellitus only; BMI, body mass index. aAdjusted for smoking history, alcohol consumption, and BMI.

REFERENCES

• 1. Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2021;71:209-49.ArticlePubMedPDF
• 2. Han E, Song SO, Kim HS, Son KJ, Jee SH, Cha BS, et al. Improvement in age at mortality and changes in causes of death in the population with diabetes: an analysis of data from the Korean National Health Insurance and Statistical Information Service, 2006 to 2018. Endocrinol Metab (Seoul) 2022;37:466-74.ArticlePubMedPMCPDF
• 3. Kang MJ, Jung KW, Bang SH, Choi SH, Park EH, Yun EH, et al. Cancer statistics in Korea: incidence, mortality, survival, and prevalence in 2020. Cancer Res Treat 2023;55:385-99.ArticlePubMedPMCPDF
• 4. SEER*Explorer. An interactive website for SEER cancer statistics. Available from: https://seer.cancer.gov/statistics-network/explorer (cited 2025 Jan 13).
• 5. Huxley R, Ansary-Moghaddam A, Berrington de Gonzalez A, Barzi F, Woodward M. Type-II diabetes and pancreatic cancer: a meta-analysis of 36 studies. Br J Cancer 2005;92:2076-83.ArticlePubMedPMCPDF
• 6. Gullo L, Pezzilli R, Morselli-Labate AM; Italian Pancreatic Cancer Study Group. Diabetes and the risk of pancreatic cancer. N Engl J Med 1994;331:81-4.ArticlePubMed
• 7. Chow WH, Gridley G, Nyren O, Linet MS, Ekbom A, Fraumeni JF, et al. Risk of pancreatic cancer following diabetes mellitus: a nationwide cohort study in Sweden. J Natl Cancer Inst 1995;87:930-1.ArticlePubMed
• 8. Malka D, Hammel P, Maire F, Rufat P, Madeira I, Pessione F, et al. Risk of pancreatic adenocarcinoma in chronic pancreatitis. Gut 2002;51:849-52.ArticlePubMedPMC
• 9. Raimondi S, Lowenfels AB, Morselli-Labate AM, Maisonneuve P, Pezzilli R. Pancreatic cancer in chronic pancreatitis; aetiology, incidence, and early detection. Best Pract Res Clin Gastroenterol 2010;24:349-58.ArticlePubMed
• 10. Kirkegard J, Mortensen FV, Cronin-Fenton D. Chronic pancreatitis and pancreatic cancer risk: a systematic review and meta-analysis. Am J Gastroenterol 2017;112:1366-72.ArticlePubMed
• 11. Munigala S, Almaskeen S, Subramaniam DS, Bandi S, Bowe B, Xian H, et al. Acute pancreatitis recurrences augment longterm pancreatic cancer risk. Am J Gastroenterol 2023;118:727-37.ArticlePubMed
• 12. Sadr-Azodi O, Oskarsson V, Discacciati A, Videhult P, Askling J, Ekbom A. Pancreatic cancer following acute pancreatitis: a population-based matched cohort study. Am J Gastroenterol 2018;113:1711-9.ArticlePubMedPDF
• 13. Kirkegard J, Cronin-Fenton D, Heide-Jorgensen U, Mortensen FV. Acute pancreatitis and pancreatic cancer risk: a nationwide matched-cohort study in Denmark. Gastroenterology 2018;154:1729-36.ArticlePubMed
• 14. Canto MI, Harinck F, Hruban RH, Offerhaus GJ, Poley JW, Kamel I, et al. International Cancer of the Pancreas Screening (CAPS) consortium summit on the management of patients with increased risk for familial pancreatic cancer. Gut 2013;62:339-47.ArticlePubMed
• 15. Aslanian HR, Lee JH, Canto MI. AGA clinical practice update on pancreas cancer screening in high-risk individuals: expert review. Gastroenterology 2020;159:358-62.ArticlePubMed
• 16. Liao KF, Lai SW, Li CI, Chen WC. Diabetes mellitus correlates with increased risk of pancreatic cancer: a population-based cohort study in Taiwan. J Gastroenterol Hepatol 2012;27:709-13.ArticlePubMed
• 17. Cho J, Scragg R, Petrov MS. Postpancreatitis diabetes confers higher risk for pancreatic cancer than type 2 diabetes: results from a nationwide cancer registry. Diabetes Care 2020;43:2106-12.ArticlePubMedPDF
• 18. Lai HC, Tsai IJ, Chen PC, Muo CH, Chou JW, Peng CY, et al. Gallstones, a cholecystectomy, chronic pancreatitis, and the risk of subsequent pancreatic cancer in diabetic patients: a population-based cohort study. J Gastroenterol 2013;48:721-7.ArticlePubMedPDF
• 19. Choi JH, Paik WH, Jang DK, Kim MK, Ryu JK, Kim YT, et al. Acute pancreatitis increases the risk of gastrointestinal cancer in type 2 diabetic patients: a Korean nationwide cohort study. Cancers (Basel) 2022;14:5696.ArticlePubMedPMC
• 20. Lee J, Lee JS, Park SH, Shin SA, Kim K. Cohort profile: the National Health Insurance Service-National Sample Cohort (NHIS-NSC), South Korea. Int J Epidemiol 2017;46:e15.ArticlePubMed
• 21. Kim MK, Han K, Lee SH. Current trends of big data research using the Korean National Health Information Database. Diabetes Metab J 2022;46:552-63.ArticlePubMedPMCPDF
• 22. Health Insurance Review & Assessment Service; National Health Insurance Service. 2020 National Health Insurance Statistical Yearbook. Wonju: HIRA, NHIS; 2020.
• 23. Seong SC, Kim YY, Khang YH, Heon Park J, Kang HJ, Lee H, et al. Data resource profile: the National Health Information Database of the National Health Insurance Service in South Korea. Int J Epidemiol 2017;46:799-800.PubMed
• 24. Sharma A, Smyrk TC, Levy MJ, Topazian MA, Chari ST. Fasting blood glucose levels provide estimate of duration and progression of pancreatic cancer before diagnosis. Gastroenterology 2018;155:490-500.e2.ArticlePubMed
• 25. Munigala S, Kanwal F, Xian H, Scherrer JF, Agarwal B. Increased risk of pancreatic adenocarcinoma after acute pancreatitis. Clin Gastroenterol Hepatol 2014;12:1143-50.e1.ArticlePubMed
• 26. Lee YH, Han K, Ko SH, Ko KS, Lee KU; Taskforce Team of Diabetes Fact Sheet of the Korean Diabetes Association. Data analytic process of a nationwide population-based study using National Health Information Database Established by National Health Insurance Service. Diabetes Metab J 2016;40:79-82.ArticlePubMedPMCPDF
• 27. Petrov MS, Yadav D. Global epidemiology and holistic prevention of pancreatitis. Nat Rev Gastroenterol Hepatol 2019;16:175-84.ArticlePubMedPMCPDF
• 28. Shen HN, Yang CC, Chang YH, Lu CL, Li CY. Risk of diabetes mellitus after first-attack acute pancreatitis: a national population-based study. Am J Gastroenterol 2015;110:1698-706.ArticlePubMedPDF
• 29. National Institutes of Health, National Institute on Alcohol Abuse and Alcoholism. Alcohol’s effects on health: researchbased information on drinking and its impact. Available from: https://www.niaaa.nih.gov/alcohol-health/overview-alcoholconsumption/moderate-binge-drinking (cited 2025 Jan 13).
• 30. Weyer C, Bogardus C, Mott DM, Pratley RE. The natural history of insulin secretory dysfunction and insulin resistance in the pathogenesis of type 2 diabetes mellitus. J Clin Invest 1999;104:787-94.ArticlePubMedPMC
• 31. Butler AE, Janson J, Bonner-Weir S, Ritzel R, Rizza RA, Butler PC. Beta-cell deficit and increased beta-cell apoptosis in humans with type 2 diabetes. Diabetes 2003;52:102-10.PubMed
• 32. Colmers IN, Bowker SL, Tjosvold LA, Johnson JA. Insulin use and cancer risk in patients with type 2 diabetes: a systematic review and meta-analysis of observational studies. Diabetes Metab 2012;38:485-506.ArticlePubMed
• 33. Lee HS, Chae W, Sung MJ, Keum J, Jo JH, Chung MJ, et al. Difference of risk of pancreatic cancer in new-onset diabetes and long-standing diabetes: a population-based cohort study. J Clin Endocrinol Metab 2023;108:1338-47.ArticlePubMedPDF
• 34. Hart PA, Bradley D, Conwell DL, Dungan K, Krishna SG, Wyne K, et al. Diabetes following acute pancreatitis. Lancet Gastroenterol Hepatol 2021;6:668-75.ArticlePubMedPMC
• 35. Zhi M, Zhu X, Lugea A, Waldron RT, Pandol SJ, Li L. Incidence of new onset diabetes mellitus secondary to acute pancreatitis: a systematic review and meta-analysis. Front Physiol 2019;10:637.ArticlePubMed
• 36. Wilcox CM, Sandhu BS, Singh V, Gelrud A, Abberbock JN, Sherman S, et al. Racial differences in the clinical profile, causes, and outcome of chronic pancreatitis. Am J Gastroenterol 2016;111:1488-96.ArticlePubMedPMCPDF
• 37. Liew SZ, Ng KW, Ishak ND, Lee SY, Zhang Z, Chiang J, et al. Geographical, ethnic, and genetic differences in pancreatic cancer predisposition. Chin Clin Oncol 2023;12:27.ArticlePubMed
• 38. Partlett C, Hall NJ, Leaf A, Juszczak E, Linsell L. Application of the matched nested case-control design to the secondary analysis of trial data. BMC Med Res Methodol 2020;20:117.ArticlePubMedPMCPDF
• 39. Pannala R, Leirness JB, Bamlet WR, Basu A, Petersen GM, Chari ST. Prevalence and clinical profile of pancreatic cancerassociated diabetes mellitus. Gastroenterology 2008;134:981-7.ArticlePubMed
• 40. Hart PA, Bellin MD, Andersen DK, Bradley D, Cruz-Monserrate Z, Forsmark CE, et al. Type 3c (pancreatogenic) diabetes mellitus secondary to chronic pancreatitis and pancreatic cancer. Lancet Gastroenterol Hepatol 2016;1:226-37.ArticlePubMedPMC
• 41. Wang Y, Ni Q, Sun J, Xu M, Xie J, Zhang J, et al. Paraneoplastic β cell dedifferentiation in nondiabetic patients with pancreatic cancer. J Clin Endocrinol Metab 2020;105:dgz224.ArticlePubMedPDF
• 42. Pereira SP, Oldfield L, Ney A, Hart PA, Keane MG, Pandol SJ, et al. Early detection of pancreatic cancer. Lancet Gastroenterol Hepatol 2020;5:698-710.ArticlePubMedPMC
• 43. Lucas AL, Kastrinos F. Screening for pancreatic cancer. JAMA 2019;322:407-8.ArticlePubMed
• 44. Ben Q, Xu M, Ning X, Liu J, Hong S, Huang W, et al. Diabetes mellitus and risk of pancreatic cancer: a meta-analysis of cohort studies. Eur J Cancer 2011;47:1928-37.ArticlePubMed
• 45. Cho SW, Kim JH, Choi HS, Ahn HY, Kim MK, Rhee EJ. Big data research in the field of endocrine diseases using the Korean National Health Information Database. Endocrinol Metab (Seoul) 2023;38:10-24.ArticlePubMedPMCPDF
• 46. Singh A, Aggarwal M, Garg R, Stevens T, Chahal P. Post-pancreatitis diabetes mellitus: insight on optimal management with nutrition and lifestyle approaches. Ann Med 2022;54:1776-86.ArticlePubMedPMC

Figure & Data

References

Citations

Citations to this article as recorded by  

About this article

Kim Ye, Yu MH, Nam CM, Kang ES. Effects of Pancreatitis and Type 2 Diabetes Mellitus on the Development of Pancreatic Cancer: A Nationwide Nested Case-Control Study. Diabetes Metab J. 2025;49(2):252-263.

Received: May 30, 2024; Accepted: Oct 23, 2024

DOI: https://doi.org/10.4093/dmj.2024.0277.

Download citation ↓