{
  "abstract": "Introduction This study aimed to identify factors associated with pancreatic abnormal findings on imaging (PAI) suggesting precancerous potential between slowly progressive type 1 diabetes and acute-onset type 1 diabetes.Research design and Methods The study was designed to identify factors associated with PAI using data from a nationwide cohort, the Japanese Type 1 Diabetes Database. Clinical factors, including sex, age, type of diabetes onset, diabetes duration, body mass index, and human leucocyte antigen genotypes associated with type 1 diabetes, were evaluated.Results Among 279 patients with type 1 diabetes, 95 patients who had not undergone imaging evaluations, two patients with pancreatic lipomatosis, and 15 patients with missing data were excluded. Finally, a total of 167 patients with type 1 diabetes were analyzed. Among 13 patients who were identified as PAI positive, female sex (92.3% vs 53.2%, p=0.007), slowly progressive type 1 diabetes (69.2% vs 36.4%, p=0.034), and older age were more common compared with PAI-negative cases. The multivariable logistic regression analysis revealed that female sex (OR 13.87; 95% CI 1.6 to 120.1; p=0.017), slowly progressive type 1 diabetes (OR 5.70; 95% CI 1.46 to 22.19; p=0.012), and age (OR 1.05; 95% CI 1.002 to 1.103; p=0.043) were independently associated with PAI positivity.Conclusions The findings indicate that slowly progressive type 1 diabetes and female sex are closely associated with PAI, along with age. These results suggest the need for increased clinical vigilance for pancreatic pathology in patients with slowly progressive type 1 diabetes.",
  "authors": [
    {
      "affiliations": [
        "Division of Diabetes, Metabolism and Endocrinology, Department of Medicine, Showa Medical University, Tokyo, Japan",
        "Seino Internal Medicine Clinic, Koriyama, Japan"
      ],
      "name": "Tomoyasu Fukui"
    },
    {
      "affiliations": [
        "Okinaka Memorial Institute for Medical Research, Tokyo, Japan",
        "Yokoyama Kazuya Cancer Research Institute, Tokyo, Japan"
      ],
      "name": "Tetsuro Kobayashi"
    },
    {
      "affiliations": [
        "Okinaka Memorial Institute for Medical Research, Tokyo, Japan",
        "Diabetes Research Center, Japan Institute for Health Security, Tokyo, Japan"
      ],
      "name": "Takuya Awata"
    },
    {
      "affiliations": [
        "Department of Endocrinology, Metabolism and Diabetes, Kindai University Faculty of Medicine, Osaka-sayama, Japan"
      ],
      "name": "Hiroshi Ikegami"
    },
    {
      "affiliations": [
        "Department of Internal Medicine (I), Osaka Medical and Pharmaceutical University, Takatsuki, Japan"
      ],
      "name": "Akihisa Imagawa"
    },
    {
      "affiliations": [
        "Department of Endocrinology and Diabetes, School of Medicine, Saitama Medical University, Iruma, Japan"
      ],
      "name": "Yoichi Oikawa"
    },
    {
      "affiliations": [
        "Department of Diabetes and Molecular Genetics, Ehime University Graduate School of Medicine, Toon, Japan"
      ],
      "name": "Haruhiko Osawa"
    },
    {
      "affiliations": [
        "Diabetes, Thyroid and Endocrine Center, Shin-Koga Hospital, Kurume, Japan"
      ],
      "name": "Eiji Kawasaki"
    },
    {
      "affiliations": [
        "Yokoyama Kazuya Cancer Research Institute, Tokyo, Japan",
        "Division of Public Health, Center for Community Medicine",
        "and Division of Cardiovascular Medicine, Department of Medicine, Jichi Medical University, Shimotsuke, Japan"
      ],
      "name": "Masanari Kuwabara"
    },
    {
      "affiliations": [
        "Yokoyama Kazuya Cancer Research Institute, Tokyo, Japan",
        "Department of Global Agricultural Sciences, Graduate School of Agricultural and Life Sciences, The University of Tokyo, Tokyo, Japan"
      ],
      "name": "Kazuhiko Kobayashi"
    },
    {
      "affiliations": [
        "Department of Diabetes and Endocrinology, Tokyo Saiseikai Central Hospital, Tokyo, Japan"
      ],
      "name": "Takeshi Katsuki"
    },
    {
      "affiliations": [
        "Department of Internal Medicine (I), Osaka Medical and Pharmaceutical University, Takatsuki, Japan"
      ],
      "name": "Norio Kanatsuna"
    },
    {
      "affiliations": [
        "Department of Metabolic Medicine, Graduate School of Medicine, Osaka University, Suita, Japan"
      ],
      "name": "Junji Kozawa"
    },
    {
      "affiliations": [
        "Department of Diabetes, Endocrinology, and Metabolism, National Center for Global Health and Medicine, Japan Institute for Health Security, Tokyo, Japan"
      ],
      "name": "Noriko Kodani"
    },
    {
      "affiliations": [
        "Department of Endocrinology and Diabetes, School of Medicine, Saitama Medical University, Iruma, Japan"
      ],
      "name": "Akira Shimada"
    },
    {
      "affiliations": [
        "Department of Pancreatic Islet Cell Transplantation, National Center for Global Health and Medicine, Japan Institute for Health Security, Tokyo, Japan"
      ],
      "name": "Masayuki Shimoda"
    },
    {
      "affiliations": [
        "Iwate Prefectural University, Takizawa, Japan"
      ],
      "name": "Kazuma Takahashi"
    },
    {
      "affiliations": [
        "Department of Diabetes, Metabolism, and Endocrinology, International University of Health and Welfare, Ichikawa, Japan"
      ],
      "name": "Daisuke Chujo"
    },
    {
      "affiliations": [
        "Department of Diabetes and Endocrinology, Toranomon Hospital Kajigaya, Kawasaki, Japan"
      ],
      "name": "Tetsuro Tsujimoto"
    },
    {
      "affiliations": [
        "Faculty of Medicine, Department of Diabetes and Endocrinology, Graduate School of Interdisciplinary Research, University of Yamanashi, Yamanashi, Japan"
      ],
      "name": "Kyoichiro Tsuchiya"
    },
    {
      "affiliations": [
        "Department of Internal Medicine (I), Osaka Medical and Pharmaceutical University, Takatsuki, Japan"
      ],
      "name": "Jungo Terasaki"
    },
    {
      "affiliations": [
        "Iwate Medical University, Shiwa, Japan"
      ],
      "name": "Kan Nagasawa"
    },
    {
      "affiliations": [
        "Department of Endocrinology, Metabolism and Diabetes, Kindai University Faculty of Medicine, Osaka-sayama, Japan"
      ],
      "name": "Shinsuke Noso"
    },
    {
      "affiliations": [
        "Department of Endocrinology and Metabolism, Nagasaki University Hospital, Nagasaki, Japan"
      ],
      "name": "Ichiro Horie"
    },
    {
      "affiliations": [
        "Department of Diabetes, Endocrinology, and Metabolism, Kyorin University, Mitaka, Japan"
      ],
      "name": "Kazuki Yasuda"
    },
    {
      "affiliations": [
        "Department of Public Health, Kobe University Graduate School of Health Sciences, Kobe, Japan"
      ],
      "name": "Hisafumi Yasuda"
    },
    {
      "affiliations": [
        "Sakai City Medical Center, Sakai, Japan"
      ],
      "name": "Toshiaki Hanafusa"
    },
    {
      "affiliations": [
        "Department of Diabetes, Endocrinology, and Metabolism, National Center for Global Health and Medicine, Japan Institute for Health Security, Tokyo, Japan",
        "Josai Clinic, Nakano Kyoritsu Hospital, Tokyo, Japan"
      ],
      "name": "Hiroshi Kajio"
    },
    {
      "affiliations": [],
      "name": "Japanese Type 1 Diabetes Database Study (TIDE-J)"
    }
  ],
  "full_text": "WHAT IS ALREADY KNOWN ON THIS TOPIC Type 1 diabetes has been recognized for an increased risk of various cancers, with a particularly strong link to pancreatic cancer. However, clinical characteristics associated with pancreatic abnormal findings on imaging (PAI) that suggest precancerous potential have not been fully examined in patients with type 1 diabetes.WHAT THIS STUDY ADDS We clearly demonstrate that slowly progressive type 1 diabetes and female sex are independently associated with PAI, achieving the ORs of 4.91 and 15.08, respectively, in patients with type 1 diabetes.HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY Our results could lead to earlier detection and better management of potential pancreatic abnormalities, especially for those with slowly progressive type 1 diabetes and females.Introduction Diabetes is associated with an increased risk of various cancers, with a particularly strong link to pancreatic cancer. 1 2 A meta-analysis found that individuals with type 1 diabetes have two times greater relative risk of pancreatic cancer compared with those without type 1 diabetes.3 Recently, we reported prevalent precursor lesions of pancreatic ductal adenocarcinoma (PDAC) in the autopsied pancreas of patients with type 1 diabetes, most predominantly in slowly progressive type 1 diabetes.4 These pancreases had pathological findings, including pancreatic intraepithelial neoplasia (PanIN) lesions and persistent enterovirus (EV) infection.4 Therefore, pancreatic abnormal findings on imaging (PAI) suggesting precancerous potential, such as pancreatic cystic lesions, duct dilatation, glandular atrophy, and tumorous lesions,5 should receive increased attention due to their potential association with PDAC in patients with type 1 diabetes. Despite the established risk factors for PDAC, the prevalence and clinical characteristics of PAI in patients with type 1 diabetes remain inadequately explored.We aimed to investigate the frequency of PAI and clinical factors associated with the presence of PAI in two subtypes of type 1 diabetes, slowly progressive type 1 diabetes6 7 and acute-onset type 1 diabetes,8 9 according to their respective diagnostic criteria.10 11 By comparing the frequency of PAI between slowly progressive type 1 diabetes and acute-onset type 1 diabetes, this research could have significant implications for the clinical management of patients with type 1 diabetes, potentially leading to more targeted screening and monitoring strategies for PDAC risk in different subtypes of type 1 diabetes.Materials and methods This is a cross-sectional study to investigate the frequency of PAI detected by transabdominal ultrasonography (US), CT, MRI including magnetic resonance cholangiopancreatography, and endoscopic retrograde cholangiopancreatography (ERCP) in patients with type 1 diabetes mellitus enrolled in the Japanese Type 1 Diabetes Database (TIDE-J). 12 Furthermore, the study assessed the factors associated with PAI, including sex, age, diabetes duration, slowly progressive type 1 diabetes or acute-onset type 1 diabetes status, body mass index (BMI), glycemic control assessed by glycated hemoglobin (HbA1c), fasting C-peptide immunoreactivity (CPR) levels, islet autoantibody status, such as glutamic acid decarboxylase autoantibodies (GADA), insulinoma-associated antigen-2 autoantibodies (IA-2A), and zinc transporter 8 autoantibodies (ZnT8A), and human leucocyte antigen (HLA) class II genes associated with susceptibility to type 1 diabetes (eg, HLA DRB1-DQB1 haplotypes).Diagnoses of slowly progressive type 1 diabetes and acute-onset type 1 diabetes were established by the attending physicians at TIDE-J study participating institutions in accordance with the respective diagnostic criteria defined by the Japan Diabetes Society.10 11 Additionally, central verification was performed by the study committee at the National Center for Global Health and Medicine (NCGM). Briefly, slowly progressive type 1 diabetes was diagnosed based on the following two criteria: (1) presence of GADA and/or islet cell antibodies at any time during the disease course, and (2) absence of ketosis or ketoacidosis at diagnosis, with no requirement for immediate insulin therapy to correct hyperglycemia.10 In the present TIDE-J study, cases with IA-2A positivity alone were also diagnosed as slowly progressive type 1 diabetes. Acute-onset type 1 diabetes was diagnosed based on the following criteria: (1) the occurrence of diabetic ketosis or ketoacidosis within approximately 3 months after the onset of hyperglycemic symptoms (eg, thirst, polydipsia, polyuria, weight loss); and (2) the need for continuous insulin therapy from the time of diagnosis. Patients were further classified as having autoimmune or non-autoimmune acute-onset type 1 diabetes depending on the presence or absence of islet autoantibodies.11Data collection As reported previously, TIDE-J is a multicenter prospective cohort study. The original published protocol for patient inclusion criteria was that (1) the duration of type 1 diabetes be less than 5 years, and (2) the patients tested positive for one or more islet-related autoantibodies and/or had fasting CPR of less than 1.0 ng/mL. 12 However, five cases were slightly beyond this period, falling within less than 6 years. After careful review by the study committee, it was deemed appropriate to include these cases. Therefore, these five cases with a type 1 diabetes duration ranging from 5.0 to 5.7 years were incorporated into the study.In the present study, patients who underwent imaging examinations at screening of pancreatic abnormalities were included.12 The following were excluded: (1) cases of acute pancreatitis and (2) cases of pancreatic lipomatosis. In this study, pancreatic cystic lesions were defined as including intraductal papillary mucinous neoplasm (IPMN) and pancreatic cyst/pseudocyst, and pancreatic tumors were defined as including PDAC and tumor-forming pancreatitis. PAI were extracted from imaging reports.Immunoassays Sera samples obtained from the participants were stored at −80°C until use. GADA and IA-2A positivity were measured using solid-phase ELISA and/or liquid-phase radioimmunoassay (RIA), as previously reported. 13–15 The cut-off values for GADA by ELISA and RIA were 5.0 and 1.5 U/mL, respectively, while IA-2As by ELISA and RIA were 0.6 and 0.4 U/mL, respectively. ZnT8A was determined by an ELISA system, with cut-off values of 10.0 U/mL.16HbA1c and C-peptide assays HbA1c and fasting CPR levels were measured using high-performance liquid chromatography and electrochemiluminescence immunoassay, respectively.HLA typing DNA of each participant was extracted from the buffy coat fraction with a DNA Blood Mini Kit (Qiagen). The HLA genotypes were centrally analyzed using next-generation sequencing at the HLA Laboratory (Kyoto, Japan). 12 We defined the positive HLA alleles associated with acute-onset and slowly progressive type 1 diabetes as DRB1-*04:05-DQB1*04:01/DRB1-*08:02-DQB1*03:02 (DR4/8) and DRB1-*04:05-DQB1*04:01 homozygotes (DR4/4) in this study.17Statistical analysis Statistical analyses were performed using the SPSS Statistics software V.25 for Windows (IBM SPSS Statistics) and JMP Pro V.16.0 software (SAS Institute Japan, Tokyo, Japan). Data were expressed as the mean (SD) or percentages (numbers), and categorical variables were compared using Fisher’s exact test (two sided). Differences in continuous values were subjected to the t -test where appropriate or the Mann-Whitney U test. Statistical significance was defined as p values <0.05. To analyze the factors associated with PAI, multivariable logistic regression was performed, with PAI occurrence as the objective variable, and sex, age, diabetes duration, slowly progressive type 1 diabetes or acute-onset type 1 diabetes status, BMI at registration, and susceptible HLA alleles associated with type 1 diabetes as the explanatory variables.Results Baseline patient characteristic data Among 279 subjects with type 1 diabetes (slowly progressive, n=109; acute onset, n=170), this study excluded 95 participants who had not undergone imaging evaluations. In addition, two patients with pancreatic lipomatosis and 15 patients with missing data were excluded. Finally, among a total of 167 patients with type 1 diabetes, 13 (7.8%) subjects were identified as PAI positive (CT 61.5%; MRI 30.8%; US 61.5%; ERCP 15.4%) and 154 (92.2%) were identified as PAI negative (CT 39.6%; MRI 2.6%; US 67.5%; ERCP 0%) ( figure 1). Among PAI-positive cases, the proportion of females was significantly higher (92.3%) compared with PAI-negative cases (53.2%) (p=0.007). Similarly, slowly progressive type 1 diabetes was more common among PAI-positive cases (69.2%) compared with acute-onset type 1 diabetes cases (36.4%) (p=0.034). PAI-positive patients were also older (59.6±15.1 years) than PAI-negative patients (47.3±17.1 years) at diagnosis (p=0.013). Other factors, including diabetes duration, BMI, frequency of susceptible HLA alleles associated with type 1 diabetes, HbA1c at diagnosis, fasting CPR levels, and islet autoantibody status at registration, were not significantly different between PAI-positive and negative groups (table 1).Table 1Background characteristics between the presence and absence of PAIPAI (+)PAI (−)P valuePatients (n)13154Female12 (92.3)82 (53.2)0.007Age at diagnosis (years)59.6±15.147.3±17.10.013Age at registration (years)60.7±15.348.7±17.20.016Duration (years)1.07±1.151.40±1.650.35Slowly progressive type 1 diabetes*9 (69.2)56 (36.4)0.034Body mass index (kg/m2)20.4±4.021.0±3.40.57Susceptible HLA alleles associated with type 1 diabetes†1 (7.7)16 (10.4)1.00Glycated hemoglobin (HbA1c) at diagnosis (%)10.4±2.711.6±2.6‡0.12Fasting CPR levels (ng/mL)1.11±1.07§1.22±1.97¶0.85Insulin treatment (yes)12 (92.3)138 (89.6)1.00Prevalence of positive islet autoantibodies at registration GADA (+)11/13 (84.6)133/152 (87.5)0.67 IA-2A (+)8/12 (66.7)63/116 (54.3)0.55 ZnT8A (+)4/13 (30.8)58/147 (39.5)0.77Data are expressed as mean±SD or n (%).*The diagnosis of slowly progressive type 1 diabetes was made based on the diagnostic criteria in the current study.†DRB1-*04:05-DQB1*04:01/DRB1-*08:02-DQB1*03:02 (DR4/8) or DRB1-*04:05-DQB1*04:01 homozygotes (DR4/4).‡Data available for 150 patients.§Data available for 12 patients.¶Data available for 134 patients.CPR, C-peptide immunoreactivity; GADA, glutamic acid decarboxylase autoantibodies; HLA, human leucocyte antigen; IA-2A, insulinoma-associated antigen-2 autoantibodies; PAI, pancreatic abnormal findings on imaging; ZnT8A, zinc transporter 8 autoantibodies.Figure 1Study diagram showing the study selection process. ERCP, endoscopic retrograde cholangiopancreatography; HLA, human leucocyte antigen; PAI, pancreatic abnormal findings on imaging; US, transabdominal ultrasonography.Of the 65 patients with slowly progressive type 1 diabetes, 13.9% (9/65) had PAI, a significantly higher prevalence compared with those with acute-onset type 1 diabetes (3.9%, 4/102). There were no significant differences in the ratio of females between the two groups. Compared with patients with acute-onset type 1 diabetes, those with slowly progressive type 1 diabetes were significantly older and had higher BMI, lower HbA1c level and longer duration of diabetes. As expected, fasting CPR levels were significantly higher in patients with slowly progressive type 1 diabetes compared with those with acute-onset type 1 diabetes (online supplemental table 1).SP110.1136/bmjdrc-2025-005229.supp1Supplementary dataDetailing characteristics of the patients with pancreatic abnormalities Table 2 shows the characteristics of patients with type 1 diabetes with PAI.Table 2Characteristics of patients with type 1 diabetes with pancreatic abnormal findings on imagingCase No*Findings (location, size)Imaging modalityFemale(Yes/No)Heavy alcohol consumption (Yes/No)†Cigarette smoking history (Yes/No)Family history of pancreas cancer (Yes/No)Pancreas surgeryPathological diagnosis1Suspected pancreatic cancer accompanied by main pancreatic duct dilationCTYesNo‡NoPancreatectomyTumor-forming pancreatitisChronic pancreatitis2Suspected pancreatic cancerCT/ERCP/USYesNoYes (past smoker)NoPancreatectomyPancreatic cancer3Branch duct-type IPMN (body, 14.6 mm)MRI/USYes‡‡NoN/AN/A4Pancreatic cystCT/USNoNoNoNoN/AN/A5Pancreatic atrophyCT/USYesNoNo‡N/AN/A6Pancreatic cyst (body, 5 mm)USYesNoNoNoN/AN/A7Pancreatic cyst (head, 5 mm)MRI/USYesNoNoNoN/AN/A8Pancreatic cystUSYesNoNoNoN/AN/A9Pancreatic atrophy and main pancreatic dilationCT/USYesNoNoYesN/AN/A10Pancreatic cyst (body, 5 mm)MRI/USYesNoNoNoN/AN/A11Pancreatic atrophyCT/USYesNoNoNoN/AN/A12Pancreatic atrophyCT/USYes‡‡‡N/AN/A13Suspected malignant main duct-type IPMN accompanied by main pancreatic duct dilation (head, 48.0 mm)CT/ERCP/MRI/USYesNoNoNoTotal pancreatectomySuspected intraductal papillary mucinous adenocarcinoma*Cases 1–9 correspond to slowly progressive type 1 diabetes, and cases 10–13 correspond to acute-onset type 1 diabetes.†≥46 g daily alcohol (ethanol) consumption.‡Not recorded.CT, computed tomography; ERCP, endoscopic retrograde cholangiopancreatography; IPMN, intraductal papillary mucinous neoplasm; MRI, magnetic resonance imaging; N/A, not applicable; US, ultrasonography.Cases 1–9 correspond to slowly progressive type 1 diabetes, and cases 10–13 correspond to acute-onset type 1 diabetes. It is noteworthy that pancreatic resection was done in three cases with suspected pancreatic cancer. Of the nine patients with slowly progressive type 1 diabetes with PAI, 44% (4/9) had a pancreatic cyst, 22% (2/9) had pancreatic atrophy, 11% (1/9) had branch duct-type IPMN (located in the body of the pancreas; size: 14.6 mm), and 22% (2/9) had solid tumor. In the two cases with solid tumors, further investigation revealed that the former was diagnosed with tumor-forming pancreatitis with chronic pancreatitis (case 1) and the latter with PDAC (case 2). Main pancreatic dilation (>2 mm) was observed in two patients with slowly progressive type 1 diabetes (cases 1 and 9).Among patients with acute-onset type 1 diabetes with PAI, 25% (1/4) had a pancreatic cyst, 50% (2/4) had pancreatic atrophy, and 25% (1/4) had suspected malignant main duct-type IPMN (located in the head of the pancreas; size: 48.0 mm). Further evaluation led to a total pancreatectomy, and pathological examination revealed findings consistent with suspected intraductal papillary mucinous adenocarcinoma (IPMA) (case 13).Independently associated factors for pancreatic abnormalities in type 1 diabetes We compared the patient background characteristics between the presence (n=13) and absence (n=154) of PAI. In the multivariable logistic regression of PAI, we found that slowly progressive type 1 diabetes (OR 5.70; 95% CI 1.464 to 22.193; p=0.012), female sex (OR 13.87; 95% CI 1.6 to 120.1; p=0.017), and age at diagnosis (OR 1.051 per 1-year increase; 95% CI 1.002 to 1.103; p=0.043) were independently associated with the presence of PAI ( table 3). Other factors, such as diabetes duration, BMI at registration, and susceptible HLA alleles associated with type 1 diabetes, showed no significant association with PAI positivity.Table 3Associated factors for pancreatic abnormal findings on imagingOR95% CIP valueFemaleVersus male13.8731.603 to 120.0630.017Age at diagnosisPer 1-year increase1.0511.002 to 1.1030.043DurationPer 1-year increase0.8770.556 to 1.3850.57Slowly progressive type 1 diabetesVersus acute-onset type 1 diabetes5.7011.464 to 22.1930.012Body mass indexPer 1 kg/m2 increase0.9250.755 to 1.1340.45Susceptible HLA alleles associated with type 1 diabetes*Positive versus negative0.5450.046 to 6.5200.63*DRB1-*04:05-DQB1*04:01/DRB1-*08:02-DQB1*03:02 (DR4/8) or DRB1-*04:05-DQB1*04:01 homozygotes (DR4/4).CI, Confidence interval; HLA, human leucocyte antigen; OR, Odds ratio.Discussion To the best of our knowledge, this is the first nationwide surveillance to examine the prevalence of PAI in patients with type 1 diabetes and clarify the clinical characteristics in these patients. In the present study, we clearly demonstrated that slowly progressive type 1 diabetes and female sex are independently associated with PAI, achieving the ORs of 5.7 and 13.87, respectively, in patients with type 1 diabetes.In the present study, pancreatic cystic lesions were identified in 7.7% (5/65) of patients with slowly progressive type 1 diabetes (IPMN, n=1; pancreatic cysts, n=4), a higher prevalence compared with 1.9% (2/102) of patients with acute-onset type 1 diabetes (suspected IPMA, n=1; pancreatic cyst, n=1). IPMN and PanIN are mucinous cystic lesions arising from pancreatic ducts and each well-recognized PADC precursor.18 Our pancreatic exocrine histology has shown that a high proportion of slowly progressive type 1 diabetes pancreas (diabetes duration, 13±7 years) had pancreatic cystic lesions associated with PanIN in the branches and smaller ducts,19 whereas acute-onset type 1 diabetes showed mild periductal fibrosis with less pancreatic ductal morphological changes.20 Additionally, the number of PanIN-positive lobes in the exocrine pancreas increased with increased duration of slowly progressive type 1 diabetes,4 21 which suggests a progressive increase of pancreatic cystic lesions associated with PanIN along with longer duration of diabetes. These results indicate that slowly progressive type 1 diabetes may be more likely to be associated with mucinous cystic precursor lesions relevant to PDAC than acute-onset type 1 diabetes. Recently, we reported that EV encoded-capsid protein 1 was detected in the exocrine as well as endocrine glands of patients with slowly progressive type 1 diabetes with varying duration of diabetes, and the persistent EV infection triggers exocrine pancreatic inflammation and transforms acinar cells into acute-onset type 1 diabetes and PanIN, followed by pancreatic cystic lesions.4 Therefore, a chronic inflammatory condition of the pancreas caused by EV infection may play some roles in the development of pancreatic cystic lesions in slowly progressive type 1 diabetes. Previous research has shown that obesity and/or insulin resistance are correlated with the development of pancreatic cysts in patients with type 2 diabetes,22 yet the mechanistic basis for these relationships is not well understood. However, all patients with pancreatic cysts were less obese. Therefore, the underlying mechanisms responsible for the development of pancreatic cysts might be different between slowly progressive type 1 diabetes and type 2 diabetes. Since our results are derived from a dataset including subjects with a relatively short disease duration, the prevalence of pancreatic cystic lesions, namely PAI, might increase over time in patients with slowly progressive type 1 diabetes.In this study, important clinical characterization of coexistence of PAI indicating precancerous potential was almost observed in females. To date, the association between type 1 diabetes and pancreatic cancer is not well described. A meta-analysis by Carstensen et al23 showed that the HRs of pancreatic cancer were 1.53 (95% CI 1.30 to 1.79) in males and 1.25 (95% CI 1.02 to 1.53) in females, respectively, indicating non-sex-specific cancers, in patients with type 1 diabetes. Interestingly, it has been reported that the majority of type 1 diabetes subjects with PADC were female by case study in a Japanese population.24 However, the pathophysiological mechanism underlying the higher incidence of PADC—dominant in females—and pronounced gender difference in the frequency of PAI observed in the present study remains unsolved. Therefore, further studies are needed to assess the exact incident rate of PAI including PADC in patients with type 1 diabetes, especially focusing on cases of slowly progressive type 1 diabetes.Current evidence strongly supports the view that the presence of one or more islet autoantibodies reflects an autoimmune process that precedes and contributes to the development of type 1 diabetes, rather than being a secondary phenomenon resulting from pancreatic damage.25 26 As shown in table 1, there was no significant difference in the positivity rate of islet autoantibodies between the PAI-positive and PAI-negative groups. These findings suggest that the presence of islet autoantibodies is associated with islet-specific autoimmune responses, while PAI may occur independently of such autoimmunity. Further investigations are needed to determine whether the coexistence of PAI affects the islet autoimmunity.Lastly, the comparisons of combinations of characteristics revealed that female sex and slowly progressive type 1 diabetes are associated with PAI, indicating precancerous potential in patients with type 1 diabetes. Furthermore, pancreatic imaging should be performed at the time of diagnosis in female subjects with slowly progressive type 1 diabetes, and those with PAI should undergo closer monitoring to facilitate early detection of PADC throughout the disease process.There were several limitations that should be addressed. First, the number of patients was relatively small. Therefore, additional studies with larger numbers of participants with type 1 diabetes will be needed to confirm our results. Second, since conventional imaging modalities of the abdomen, such as US, CT, or MRI, were used as a screening test in this study, the prevalence of PAI may have been underestimated due to the low sensitivity for identifying pancreatic cystic lesions compared with endoscopic US and/or ERCP. The choice of imaging modality was determined by the diabetes specialist at each facility, and no standardized criteria were used. Although imaging was performed in all cases, the variety of modalities used may also represent a limitation. Third, several population-based imaging studies have reported that pancreatic findings, such as cystic lesions and glandular atrophy, increase with age. Some studies have also shown a slightly higher prevalence of these findings in women, even among the general population or individuals without known pancreatic disease.27 28 These observations may be related to physiological ageing processes or sex-related anatomical or hormonal factors. Unfortunately, our study did not include a non-diabetic control group with imaging data for direct comparison. Therefore, although PAI findings were more frequent in elderly individuals and women with type 1 diabetes in the current study, these results may partially reflect general population trends, rather than being specific to type 1 diabetes. Further studies are needed to clarify the role of ageing and sex differences in PAI development in patients with type 1 diabetes.In conclusion, concomitant PAI may be a distinct clinical feature of slowly progressive type 1 diabetes, compared with acute-onset type 1 diabetes.",
  "title": "Slowly progressive type 1 diabetes and female sex as associated factors for pancreatic abnormalities on diagnostic imaging indicating precancerous potential",
  "uid": "b53504dd-8fc2-539b-b22d-7634651ada6e"
}