Management pankreatisch-zystischer Neoplasien und Validität der aktuellen Leitlinien; eine retrospektive Datenanalyse eines Patientenkollektivs des Universitätsklinikums Marburg
Loading...
Files
Date
2020-10-20
Authors
Publisher
Philipps-Universität Marburg
item.page.supervisor-of-thesis
Abstract
As a result of widespread use of cross-sectional imaging, clinicians are confronted with pancreatic cysts with increasing frequency (De Jong et al. 2010). The adequate diagnosis and therapy of patients with pancreatic cystic lesions represents an everyday clinical challenge and varies considerably between the different entities of the pancreatic cystic lesions (Visser et al. 2008). Pancreatic cysts comprise of a heterogeneous group of entities. Some of which are benign such as inflammatory pseudocysts or serous cystic neoplasms (SCN) and do not require resection when asymptomatic. Others like mucinous cystic neoplasm (MCN) or intraductal papillary mucinous neoplasms (IPMN) have a malignant potential and in these cases surgical resection is often indicated. Due to the varying potential for malignant degeneration and progression to pancreatic ductal adenocarcinoma (PDAC) an early and differentiated diagnosis is crucial to ensure adequate clinical management of PCN-patients (De Jong et al. 2010).
A clinical problem is the correct preoperative identification of the respective PCN-entity on basis of image morphological criteria. However, due to the varying potential for malignant degeneration of the various PCN-entities, the correct diagnosis is decisive for the further clinical procedure (Del Chiaro et al. 2014).
In recent years, various guidelines with therapeutic recommendations for PCN-patients have been published (Tanaka et al. 2012) (Vege et al. 2015). Currently, there is no international consensus on the clinical approach in these patients (Hasan et al. 2019) (van Huijgevoort et al. 2019). So far, there was no systematic investigation of the clinical procedure in PCN-patients within the UKGM. The European Study Group (ESG) guideline on the clinical management of PCN-patients published in 2018 is the first evidence-based guideline on this topic. However, their validity has not yet been investigated in a mixed group of patients with different PCN- entities.
In this retrospective study, a group of 219 patients with different PCN were examined with regard to their clinical management. The data provide a comprehensive picture of the clinical procedure. The majority of PCN-patients (81.7 %) was asymptomatic at initial diagnosis. The most common symptom was abdominal pain. 16.0 % of the study population had PCN related pain.
85.4 % of the patients received conservative therapy primary. Symptomatic patients tend to be treated surgically more frequently than patients without symptoms. The most common radiological diagnosis was BD-IPMN with 86.2 %. The other PCN-entities were diagnosed less frequently (MD/MT IPMN 4.6 %, MCN 1.4 %, SCN 7.8 %).
The specificity of radiological methods for preoperative PCN-entity determination was 77.8 %.
The overall prevalence of malignancy in our study population was 4.1 % at the initial diagnosis. Within the surgically treated patients, however, the prevalence of malignancy varies greatly depending on the PCN-entity (BD-IPMN 20.0 %, MD/MT-IPMN 50.0 %, MCN 0.0 %, SCN 0.0 %).
2.7 % of these primarily conservatively treated patients developed a progression in the sense of the new occurrence of ESG risk criteria during observation. Four patients (1.8 % of the study population) received secondary surgical treatment. One of the four patients showed a BD-IPMN associated carcinoma. This corresponds to a malignancy incidence of 0.53 % of all PCN-patients treated primarily conservatively and 0.56 % of BD-IPMN patients treated primarily conservatively within 52 months. In the secondary operated patient, with evidence of a BD-IPMN associated carcinoma, the progression was detected after monitoring of 52 month.
Overall, 16.4 % of the study group was treated surgically. Only 27.8 % of the surgically treated patients (4.6 % of the study population) showed malignancy at histopathological examination. Accordingly, no malignancy was detected in 72.2 % of surgically treated patients (11.8 % of the study population).
Progression-free survival at 54 months was 98.8 % in the group without ESG risk criteria in the initial imaging. Progression-free survival in the group of patients with relative indication of surgery was 92.9 % at 54 months. In the group of patients with absolute indication of surgery in initial imaging, progression-free survival after 54 months was 40.0 %. The differences in progression-free survival were statistically significant.
The ESG-risk-criteria were tested for statistical validity in predicting malignancy. Some risk criteria showed a association with the presence of malignancy. No association could be established for other ESG-criteria. The presence of solid cysts mass showed the strongest association with the presence of malignancy (odds ratio= 7.77). In addition enhancing murale nodules ≥ 5 mm (odds ratio= 6.25) and < 5 mm (odds ratio= 6.30) showed a strong association with presence of malignancy. Main pancreatic duct dilatation of ≥ 10 mm as well as the dilatation between 5 and 9. 9 mm showed an association with presence of malignancy (odds ratio= 2.17 and 2.19). An association with malignancy (odds ratio= 3.00) was also demonstrated for an elevated serum CA 19-9 level of ≥ 37 U/ml. A cyst diameter of ≥ 4 cm showed no association (odds ratio= 0.30).
With regard to the detection of malignancy, the individual ESG-criteria of absolute indication for surgery showed a high specificity (76.9 %- 100 %), but a low sensitivity (10.0 % -70.0 %). The individual ESG-criteria of relative indication for surgery also showed high specificity
(73.1 %- 94.7 %), but low sensitivity (10.0 %- 60.0 %).
Through a combination of risk criteria, the sensitivity could be increased in particular. The presence of at least one criterion of relative indication for surgery showed a sensitivity of
90.0 %. The highest specificity (100 %) was achieved by combining at least one absolute and two additional relative risk criteria.
Most PCN are found incidentally in asymptomatic patients. The BD-IPMN is the PCN-entity with the highest prevalence. It is obvious that the relative frequencies of the different PCN-entities in surgical case series are subject to a selection bias, since the different PCN are resected with varying strategies due to their different malignant potential and the varying frequency of occurrence of risk criteria. The prevalence of malignancy is low in our study population at the initial diagnosis with 4.1 %. However, this varies greatly depending on the respective PCN-entity. The different prevalences of malignancy in resected MD/MT-IPMN and BD-IPMN (50 % vs. 20 %) in our data underscore the importance of correct preoperative diagnosis to assess the risk of malignancy. The low specificity of 77.8 % with regard to preoperative entity determination by radiological methods underscores the need for further studies in order to optimize preoperative imaging diagnostics with regard to correct PCN determination.
Overall, the prevalence of malignancy was low in the group of operated patients at 27.8 %. No malignancy was detected in 72.2 % of surgically treated patients. Retrospectively, these patients were treated unnecessarily surgically from an oncological point of view. The aim should be to further improve the preoperative identification of patients with a malignant disease to minimize the rate of unnecessarily operated patients.
The incidence of malignancy in the primary conservative treated group is low within 54 months. However, progression can occur after longer period of time. The monitoring periods in our study are not sufficient for a final assessment of the progression rate within the case group of primarily conservatively treated patients. A further long-term observation of the study population is necessary in order to assess the progression rate more accurately. Overall, however, the collected data support the recommendation of the ESG to follow up IPMN- patients as long as they are fit for surgery.
Our data indicate that patients with risk criteria in the initial imaging have a higher probability of progression and thus the development of a carcinoma than patients without risk criteria.
These data support the recommendation in the ESG guideline to intensify the monitoring of patients with criteria of relative indication of surgery in the initial imaging in individual cases where primary surgery is not to be forced due to an increased perioperative risk.
Our data underscore the importance of solid cyst mass and murale nodules as most important risk criteria with high relevance for the preoperative prediction of malignancy. Dilatation of the main pancreatic duct and elevated serum CA 19-9 levels are also important criteria for assessing the likelihood of malignancy. The data suggest that the risk criteria listed in the ESG guideline are an important tool for the preoperative prediction of malignancy even within a mixed PCN-collective. However, not every single criterion should be considered in isolation, but all ESG risk criteria should always be taken into account. The available data suggest that the above combinations of risk criteria can improve diagnostic accuracy. The extent to which the above combinations of risk criteria should be used in clinical practice to improve preoperative diagnostic accuracy requires further prospective testing. It would be conceivable and fundamentally desirable to develop a preoperative risk score to assess the likelihood of malignancy in PCN-patients. However, the data available to us are not yet sufficient to develop such a score. Further long-term studies are needed to generate a more comprehensive understanding of the biological behaviour of PCN and to improve the clinical management of PCN-patients.
Review
Metadata
Contributors
Supervisor:
Dates
Created: 2020Issued: 2020-10-20Updated: 2020-10-20
Faculty
Medizin
Language
ger
Data types
DoctoralThesis
DDC-Numbers
610
show more
Zumblick, Malte: Management pankreatisch-zystischer Neoplasien und Validität der aktuellen Leitlinien; eine retrospektive Datenanalyse eines Patientenkollektivs des Universitätsklinikums Marburg. : Philipps-Universität Marburg 2020-10-20. DOI: https://doi.org/10.17192/z2020.0376.