Introduction

Defining pancreatic steatosis (PS) and establishing its association with pancreatic cancer has been a discussed theme in the last decade. However, no consensus guidelines have yet been published on the clinical relevance and management of PS. In this systematic review and meta-analysis, we investigated the association between PS and pancreatic ductal adenocarcinoma (PDAC).

Aims & Methods

Medical literature between 2007 and 2023 was reviewed for eligible trials investigating the prevalence of PS in patients with PDAC. Eligible studies reporting on PS, assessed via imaging or histology, were included. Our primary objective was to determine the association between PDAC and PS by comparing the prevalence of PS in individuals with- and without PDAC. Subsequently, an evaluation was conducted to determine whether the method of assessment (CT, MRI, EUS or histology) and study country was correlated with the association between PS and PDAC. Measures of effect size were determined using odds ratios (ORs) and corresponding 95% confidence intervals (95% CI).

Results

The systematic review identified a total of 23 studies, of which seventeen studies examined PS prevalence among PDAC patients. Overall, the pooled prevalence of PS in patients with PDAC. was 53.6% (95% CI 40.9-66.2). No significant difference in PS prevalence was observed across various diagnostic methods or geographical regions (p= 0.71 and p= 0.50, respectively). Overall, the pooled OR for PS in in patients with PDAC compared to controls was 3.23 (95% CI 1.86-5.60, Figure 1). Subgroup analysis did not reveal any difference according to country (p= 0.52), but higher OR was observed in studies using EUS for the assessment of PS (pooled OR=6.26, 95% CI 1.23-31.8) compared to studies using CT (pooled OR=3.31, 95% CI 1.93-5.68), histology (pooled OR=1.76, 95% CI 1.22-2.54) or MRI (pooled OR=1.19, 95% CI 1.06-1.33).

Conclusion

Patients with PDAC have notable prevalence of pancreatic steatosis (53.6%). Furthermore, PDAC patients are significantly more likely to have PS (OR 3.23). These findings underscore the importance of prioritizing a standardized approach to the diagnosis, follow-up, and treatment of PS. Future studies should prioritize the identification of patients with PS who will benefit from PDAC surveillance programs.

Introduction

Incidence of sporadic ductal pancreatic adenocarcinoma (PDAC) in the Czech Republic is the third highest in Europe (23.1/100,000 population). New-onset diabetes mellitus with weight loss (> 5% within < 6 months) in patients over sixty can be an early sign of PDAC. New-onset diabetes mellitus was recorded in 12.2% subjects three years prior to the diagnosis of PDAC (in the 60-74+ age group:13.8%) in the Czech Republic in the period 2017-2021 [ref. 1].

Aims & Methods

The aim of this study was to evaluate the standardized incidence ratio (SIR) of PDAC in patients with diabetes mellitus with and without previous history of any other malignancy within a 12-year period.
Methods: The analysis was based on individual population data from the Czech National Registry of Reimbursed Health Services, the National Cancer Registry, and the Death Certificate Information System in the period 2010 - 2021. Patients were divided into two groups: the first one included subjects with diabetes mellitus who had a history of any malignancy prior to the diagnosis of diabetes (follow-up 865,956 person-years); the second cohort consisted of people with diabetes mellitus who had no history of malignancy prior to the diagnosis of diabetes (7,971,090 person-years). The calculated person-years at risk were stratified by sex, age and time periods (2010-2013; 2014-2017; 2018-2021). Subgroup analyses were performed considering follow-up and age at diagnosis of diabetes. The first report of any antidiabetic drug (ATC group A10) was the determinant for the diagnosis of diabetes mellitus in a particular individual. SIR was calculated as the ratio of observed to expected (estimated from incidence in the general population) cases of PDAC.

Results

Selected SIRs stratified according to age and follow-up are displayed in Table 1.
A significantly higher risk of PDAC in diabetes mellitus with a history of previous malignant diseases (compared to the general population) was found for the following: head and neck cancer (SIR 2.54), gastric cancer (2.12), colorectal cancer (1.38), laryngeal cancer (2,51), malignant melanoma (1.65), breast cancer (1.61), cervical cancer (2.22), uterine cancer (1.72), prostate cancer (1.59), testicular cancer (3.1), kidney cancer (1.75), urinary bladder cancer (1.74), thyroid cancer (2.13), Hodgkin lymphoma (2.74), non-Hodgkin lymphoma (2.15) and leukaemias (1.93).

Table 1
Risk of PDAC in diabetes mellitus

Diabetes mellitus with history of previous malignant disease n = 144,946 individuals		Diabetes mellitus with no history of malignant disease n = 1,123,065 individuals
Age at diagnosis of diabetes mellitus	SIR (95% C.I.)	Age at diagnosis of diabetes mellitus	SIR (95% C.I.)
60 - 69 years	3.13 (2.84 - 3.45)	60 - 69 years	2.12 (2.04 - 2.19)
70 - 79 years	1.58 (1.45 - 1.73)	70 - 79 years	1.05 (1.01 - 1.10)
≥ 80 years	0.79 (0.68 - 0.91)	≥ 80 years	0.63 (0.58 - 0.68)
Follow-up	SIR (95% C.I.)	Follow-up	SIR (95% C.I.)
up to 3 months	5.49 (4.62 - 6.49)	up to 3 months	4.82 (4.46 - 5.20)
3 months to 1 year	2.79 (2.41 - 3.20)	3 months to 1 year	2.49 (2.34 - 2.65)
from 1 to 2 years	1.74 (1.48 - 2.04)	from 1 to 2 years	1.62 (1.52 - 1.74)

C.I.: Confidence Intervals

Conclusion

Risk of PDAC in patients with diabetes mellitus was higher in subjects with a history of various malignant diseases compared to those with no previous history of any malignancy. Particular attention should be paid to new-onset diabetes (diagnosis made within one last year). History of a malignancy, as an important risk factor for PDAC, should be taken into account when surveilling patients with diabetes mellitus.

Funding: Research projects MO1012 (DZVRO) and MH CZ-DRO (UHHK, 00179906).

References

Ref. 1: UEG Journal 2024; 12 (Suppl S8): 565.

Disclosure

There is conflict of interest.

Introduction

Type 1 AIP is the most common presentation of IgG4-related disease. Serum IgG4 levels are widely used in initial screening and diagnosis, with levels ≥135 mg/dL recommended as part of the diagnostic criteria. However, its diagnostic specificity is limited as it can be elevated in a broad spectrum of neoplastic, infectious, and autoimmune diseases. Histological diagnosis though definitive, remains challenging in real life clinical practice. In recent years, IgG4/IgG ratio has been proposed as a promising alternative serum biomarker.

Aims & Methods

We aimed to study the diagnostic utility of serum IgG4/IgG ratio compared to serum IgG4 alone in the diagnosis of Type 1 AIP. We identified a cohort of patients from 2015 to 2021 who had raised serum IgG4 levels in Tan Tock Seng Hospital, Singapore. Associations between serum IgG4 level and IgG4/IgG ratio were compared to confirmed AIP diagnosis and characteristic computed tomography (CT) and magnetic resonance imaging (MRI) features (diffuse pancreatic swelling, rim-like enhancement). Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated at IgG4/IgG thresholds of ≥15% and ≥20%.

Results

Our sample included a total of 60 patients with raised serum IgG4 levels. 21 patients had been diagnosed with Type 1 AIP, and 39 patients had not.
The majority of our patients had serum IgG4 ≥135 mg/dL (n=53, 88.3%). However, there was no significant association between the conventional serum IgG4 cut off of ≥135 mg/dL and the diagnosis of AIP (p = 0.687). In contrast, serum IgG4/IgG ratio demonstrated significantly higher values in patients with AIP compared to those without (median 23.2% vs 13.9%; p < 0.001). Analysis by logistic regression of IgG4/IgG ratio showed an acceptable performance with an area under the curve of 0.779, sensitivity of 92.3%, and specificity of 52.4%.

Variable	Type 1 AIP Present (n=21)	Type 1 AIP Absent (n=39)	Test Statistic	p-value
IgG4/IgG ratio	23.2 (24.0 ± 14.2)	13.9 (13.4 ± 4.5)	U = 184	<0.001
Serum IgG4 ≥135 mg/dL	18/21 (85.7%)	35/39 (89.7%)	Fisher's exact	0.687

We performed further analysis to determine a threshold for diagnostic utility of serum IgG4/IgG ratio. A ratio of ≥20% was strongly associated with AIP diagnosis (OR = 13.2, 95% CI: 3.08–56.5; p < 0.001), with high specificity (92.3%) and PPV (78.6%). Reducing the diagnostic threshold to ≥15% increased sensitivity to 76.2% (vs 52.4% at ≥20%) and improved NPV 84.4%, however this reduced specificity to 69.2%.

Subset analysis was performed among patients with Type 1 AIP (n=21) to evaluate for relationship between serum biomarkers and typical imaging findings (diffuse pancreatic swelling, rim-like enhancement). When stratified to a cut off of ≥135 mg/dL, the association between serum IgG4 levels with imaging findings trended towards positivity, but was not statistically significant (OR 3.14, p = 0.0553, φ 0.196). The IgG4/IgG ratio demonstrated a stronger association with typical imaging findings, although still not achieving statistical significance (OR = 1.065, p = 0.129). This subset analysis is likely limited by the small sample size.

Conclusion

This study demonstrates that serum IgG4/IgG ratio is a superior serological marker compared to serum IgG4 alone, particularly in the diagnosis of type 1 AIP. A threshold ratio of ≥20% demonstrated strong association with AIP diagnosis, with high specificity and PPV.
Our findings support the incorporation of IgG4/IgG ratio - particularly at a cut-off of 20% - as a valuable adjunct in diagnosis of suspected type 1 AIP, and we recommend further research in a larger cohort to validate this data.

References

1. Wallace ZS, Naden RP, Chari S, Choi HK, Della-Torre E, Dicaire JF, et al. The 2019 American College of Rheumatology/European League Against Rheumatism classification criteria for IgG4-related disease. Arthritis Rheumatol. 2020;72(1):7–19.
2. Umehara H, Okazaki K, Kawa S, Takahashi H, Yoshino T, Matsui S, et al. The 2020 revised comprehensive diagnostic (RCD) criteria for IgG4-related disease. Mod Rheumatol. 2021;31(3):529–33.
3. Lanzillotta M, Mancuso G, Della-Torre E. Advances in the diagnosis and management of IgG4-related disease. BMJ. 2020;369:m1067.
4. Uchida K, Okazaki K. Current status of type 1 (IgG4-related) autoimmune pancreatitis. J Gastroenterol. 2022;57(10):695–708.
5. Xia CS, Zhang S, Duan L, Zhao CW, Zhang F. Diagnostic performances of serum IgG4 concentration and IgG4/IgG ratio in IgG4-related disease. Clin Rheumatol. 2017;36(11):2769–74.