Effectiveness of endoscopic ultrasound-guided tissue acquisition with stereomicroscopic on-site evaluation for preoperative diagnosis of resectable or borderline resectable pancreatic cancer: a prospective study
Article information
Abstract
Background/Aims
To validate endoscopic ultrasound-guided tissue acquisition (EUS-TA) used in conjunction with stereomicroscopic on-site evaluation (SOSE) as a preoperative diagnostic tool for resectable pancreatic cancer (R-PC) and borderline resectable PC (BR-PC).
Methods
Seventy-eight consecutive patients who underwent EUS-TA for suspected R-PC or BR-PC were enrolled. The primary endpoint was the sensitivity of EUS-TA together with SOSE based on the stereomicroscopically visible white core (SVWC) cutoff value. One or two sites were punctured by using a 22-gauge biopsy needle for EUS-TA, based on the SOSE findings.
Results
We collected 99 specimens from 56 and 22 patients with R-PC and BR-PC, respectively. Based on the SOSE results, we performed 57 procedures with one puncture. The SVWC cutoff values were met in 73.7% and 73.1% of all specimens and in those obtained during the first puncture, respectively. The final diagnoses were malignant and benign tumors in 76 and two patients, respectively. The overall sensitivity, specificity, and accuracy of EUS-TA for the 78 lesions were 90.8%, 100%, and 91.0%, respectively. The sensitivity for malignant diagnosis based on the SVWC cutoff value were 89.5% and 90.4% for the first puncture and all specimens, respectively.
Conclusions
The sensitivity of EUS-TA in conjunction with SOSE for malignancy diagnosis in patients with suspected R-PC or BR-PC was 90.4%.
INTRODUCTION
Endoscopic ultrasound-guided tissue acquisition (EUS-TA) is a safe and standard procedure for pathological diagnosis of pancreatic cancer.1-4 As a preoperative adjuvant, it improves the survival of patients with resectable pancreatic cancer (R-PC) and borderline resectable PC (BR-PC).5-9 Therefore, obtaining a definitive pathological diagnosis of pancreatic cancer before surgery is crucial. However, most studies on EUS-TA for radically R-PC and BR-PC or small pancreatic masses are retrospective, and prospective studies are scarce.10-14 Improvements are required in the diagnostic yield and safety of EUS-TA for small solid pancreatic lesions; furthermore, investigations are also required into the optimal needle and technique for the procedure.15 This is particularly important because of the recently reported cases of needle tract seeding (NTS);16-18 almost all occurrences of NTS are noted to have been preceded by several punctures during EUS-TA.
Rapid on-site evaluation (ROSE),19 macroscopic on-site evaluation (MOSE),20 sample isolation processing by stereomicroscopy,21 and stereomicroscopic on-site evaluation (SOSE)22 can be performed to assess the quality of the specimens obtained on-site via EUS-TA. ROSE is the most common among these procedures, but it requires an on-site cytopathologist; this may be feasible for most hospitals in the United States, but not for approximately 50% of the hospitals in Europe and Asia.23 Conversely, SOSE does not require a cytopathologist, and specimens are diagnosed when a stereomicroscopically visible white core (SVWC) cutoff value of ≥11 mm is obtained.22 The adequacy of specimens obtained by using EUS-TA can be directly or indirectly determined via on-site evaluation, and the procedure can be terminated. Thus, the number of punctures can be minimized. However, the relevance of the SOSE in terms of R-PC and BR-PC has not yet been determined.
Therefore, this prospective interventional study on R-PC and BR-PC (at a clinical stage where radical resection is possible) aimed to verify the diagnostic SOSE findings of specimens obtained by using a minimum number of punctures during EUS-TA.
METHODS
Study design
This prospective interventional study was conducted at two medical centers. The primary endpoint was the diagnostic sensitivity of EUS-TA combined with SOSE for malignant specimens, based on the SVWC cutoff value. The secondary endpoints were the ratio of specimens meeting the cutoff values, sensitivity of EUS-TA combined with cytology and histology, specificity, positive and negative predictive values (PPV and NPV, respectively), accurate diagnosis rates (ADRs), and EUS-TA-associated adverse events occurring within 30 days of the procedure.
Patient eligibility
Patients who underwent EUS-TA-based assessment of pancreatic tumors and were suspected to have pancreatic cancer at Kitasato University Hospital and the Japan Community Health Care Organization Sagamino Hospital were enrolled between March 2021 and January 2023.
The inclusion criteria were as follows: age ≥20 years, suspicion of R-PC or BR-PC based on the General Rules for the Study of Pancreatic Cancer published by the Japan Pancreas Society (fourth English Edition), presence of pancreatic tumors that required EUS-TA for pathological diagnosis, adequate organ function, and provision of written informed consent. The exclusion criteria were allergies or renal dysfunction that contraindicated the use of iodine-containing contrast agents.
EUS-TA procedure
After conscious sedation with midazolam, EUS-TA was performed by eight endoscopists, comprising four fellows and four trainers who were board-certified by the Japan Gastroenterological Endoscopy Society. A GF-UCT260 linear scanning video echoendoscope (Olympus Medical), an EU-ME2 Premier Plus dedicated ultrasound processor (Olympus Medical), and an Acquire 22-gauge biopsy needle (Boston Scientific Corp.) were used. The endoscopists used a 20 mL syringe for aspiration, and the needle was stroked approximately 20 times within the lesion. The procedure was terminated when the SVWC cutoff was met following a single puncture or when the maximum number of punctures was two (irrespective of whether the SVWC cutoff was met). The incidence of adverse events within 30 days of EUS-TA was evaluated in outpatient clinics.
SOSE
Two designated evaluators (K.O. and M.W.) conducted the SOSE by using protective gloves, glasses, and clothing in a well-ventilated environment. The specimens obtained via EUS-TA were evaluated by using a stereomicroscope under a magnified field of view, as described previously.22 Both of these evaluators had also participated as evaluators in a previous study.22 As a rule, SVWCs and blood clots were differentiated based on coloration under the stereomicroscope. Furthermore, brittle white specimens identified by using injection needles were considered necrotic tissues rather than SVWCs. Tissue sections obtained by using EUS-TA were immersed in 10% neutral-buffered formalin and submitted to the pathology laboratory.
Pathological diagnosis
Pathologists stained the specimens with hematoxylin and eosin and Papanicolaou hematoxylin for histological and cytological diagnoses, respectively.
In patients treated surgically during the observation period, the diagnosis was considered valid if the pathological findings of the surgical specimen were consistent with those obtained using EUS-TA. In cases where tumors were deemed unresectable after EUS-TA, we first confirmed whether the clinical course and imaging findings for >6 months from the date of EUS-TA were consistent with the pathological diagnosis obtained with EUS-TA and then determined the validity of the diagnosis.
Statistical analysis
Based on a previous study, we assumed that the frequency of meeting the SVWC cutoff value would be 66.7% and that the sensitivity of a malignant diagnosis based on that value would be 94.4%.22 We determined that 65 specimens were required to satisfy the SVWC cutoff at the first puncture (one-sided α, 0.05). We also considered a detection power of 80% and an equivalence tolerance margin of 10% for the sensitivity of the malignant diagnosis. Hence, 74 patients were required to obtain the required number of specimens, with up to two punctures. Assuming a dropout rate of approximately 5%, the target number of patients was set to 78.
The presence of SVWC and the tissue sampling rate were classified as positive if white samples were visible by using stereomicroscopy and if the lesion tissues were visible by using optical microscopy. Accuracy was based on combined cytological and histological diagnoses.
The 95% confidence interval (CI) was calculated by analyzing differences in population ratios. Categorical variables were compared using Fisher’s exact probability tests, and values with two-tailed p<0.05 were considered statistically significant. Data were statistically analyzed using R ver. 3.2.4 (R Foundation for Statistical Computing) and BellCurve for Excel ver. 4.03 (Social Survey Research Information Co., Ltd.).
Ethical statements
This study complied with the principles of the Declaration of Helsinki (2013 amendment) and was approved by the Ethics Review Boards of the Kitasato University School of Medicine and the Japan Community Health Care Organization Sagamino Hospital on the basis of ethical, scientific, and medical validity (approval numbers: C21‒018 and 202101). All the patients provided written informed consent for participation. This study was registered at http://www.umin.ac.jp (UMIN 000044023).
RESULTS
Patient characteristics
Table 1 presents patient characteristics (median age, 73 [40‒85] years; men, n=45 [57.7%]). Overall, 47 of the 78 patients had pancreatic head tumors (60.3%). The median maximum diameter of the pancreatic tumors was 19 (4–45) mm, and the maximum tumor diameter was <20 mm in 42 patients (53.8%). Among these tumors, 56 (71.8%) were R-PC and 22 (28.2%) were BR-PC tumors. Additionally, BR-PC invaded the portal vein and abutted the major arteries in 16 and six patients, respectively. The final diagnoses were pancreatic ductal adenocarcinoma (n=74), neuroendocrine neoplasm (n=2), tumor-forming pancreatitis (n=1), and autoimmune pancreatitis (n=1).
Characteristics of the patients and lesions (n=78)
EUS-TA procedure
Table 2 shows the results of EUS-TA. The technical success rate was 100% and 99 punctures were required for 78 lesions. The median number of punctures was one (1‒2), and the procedure was completed with only one puncture in 57 (73.1%) lesions, based on the SOSE results. The puncture routes were transgastric, transduodenal (descending), and transduodenal bulb in 32 (41.0%), 36 (46.2%), and 10 (12.8%) patients, respectively. No complications associated with EUS-TA were observed.
Results of endoscopic ultrasound-guided tissue acquisition
Diagnostic yields of EUS-TA with SOSE
Table 3 shows the results of rapid specimen evaluation via SOSE for the entire cohort. Overall, 73.7% of the specimens met the SVWC cutoff value. Specifically, among the 78 and 21 specimens collected during the first and second punctures, respectively, 73.1% and 76.2% met the SVWC cutoff values. The median time required for evaluation using SOSE was 32 (4‒126) seconds. The collection rate was 100% (99/99). The sensitivity, specificity, PPV, NPV, and ADR of EUS-TA were 90.8%, 100%, 100%, 22.2%, and 91.0%, respectively (Table 4). For 95 specimens from 76 lesions (excluding those from two patients diagnosed with benign tumors during the observation period), the sensitivities for malignant diagnosis at the first and second punctures based on the SVWC cutoff value were 89.5% and 93.8%, respectively; the overall sensitivity was 90.4%. The diagnosis rate did not show a significant difference between the specimens that met and those that did not meet the SVWC cutoff value (90.4% vs. 86.4%).
Assessment of the 78 stereomicroscopic on-site evaluations
Diagnostic yields of endoscopic ultrasound-guided tissue acquisition with stereomicroscopic on-site evaluation
DISCUSSION
NTS after EUS-TA was first established for lymph node metastases of melanomas.24 Thereafter, NTS was identified in pancreatic cancer.25 A recent retrospective survey of 12,109 primary pancreatic tumors across 235 facilities in Japan found an NTS incidence of 0.330%, which increased to 0.409% when limited to pancreatic cancer.16 Another retrospective study limited to pancreatic body and tail cancer found an NTS incidence of 3.4%.17 The authors concluded that preoperative EUS-TA for pancreatic body and tail cancer did not adversely affect the long-term prognosis; however, NTS developed in a small proportion of patients. NTS associated with EUS-TA typically presents with a submucosal tumor-like morphology. Therefore, its spread to the gastrointestinal wall in surgically resected pancreatic bodies and tail cancer followed up for a certain period after surgery can be proven; iatrogenic dissemination into the abdominal cavity due to EUS-TA and metastasis are otherwise difficult to differentiate. Thus, the frequency of NTS, or its dissemination, may be higher. The origin of the NTS remains unknown. However, it is speculated that malignant cells isolated upon puncturing a pancreatic tumor with a needle migrate to the gastrointestinal tract walls.26,27 Furthermore, microscopic bleeding resulting from a puncture and the consequent reactive changes may promote tumor cell survival in the gastrointestinal wall.18 Thus, NTS can be caused by a single puncture.24,28 Kurosu et al.29 prospectively investigated whether cells derived from pancreatic cancers adhered to the puncture needle’s external surface after EUS-TA and found that the rate of positive lavage cytology for the needle’s external surface was 20%. Thus, multiple punctures would naturally increase the risk of NTS. Accordingly, when postoperative NTS is a concern, EUS-TA should be performed with a minimum number of punctures.
In the present study, the sensitivity for malignant diagnosis in specimens that met the SVWC cutoff value at the first puncture was 89.5%, as compared with the 94.4% determined in a previous study.22 With a difference of 4.9% (95% CI, ‒8.37 to 18.17), the findings of the present study are inconsistent with those of the previous study. The first reason may be the small diameters of the target lesions used in this study. The aforementioned study verified the usefulness of SOSE in 58 patients with unresectable PC and 12 patients with BR-PC; the median maximum tumor diameter in that study was 35 (26–44) mm.22 The maximum tumor diameter was significantly smaller in the present study (p<0.01). Small tumors may affect the diagnostic performance of EUS-TA. In a retrospective analysis of EUS-TA in 159 patients with pancreatic masses, ADRs of 97% and 64% were observed when the tumor diameters were ≥10 and <10 mm, respectively.10 Moreover, in another retrospective analysis of EUS-TA in 944 patients with pancreatic masses, ADRs of 82.5%, 83.5%, and 93.4% were observed when the tumor diameters were ≤10, 10‒20, and ≥20 mm, respectively.11 The authors of that study also noted that the ADRs decreased when the lesions were <20 mm and when ROSE was not included in the multivariate analysis. In the present study on small-sized tumors of R-PC and BR-PC, EUS-TA similarly yielded diagnostic results that were inferior to those obtained by EUS-TA for unresectable PC in a previous study.22 The second reason may be that SOSE allows an indirect assessment of the adequacy of the specimens, similar to MOSE.20 The diagnostic sensitivity of SOSE was high when the SVWC cutoff value was met. However, unlike ROSE, SOSE does not allow rapid and direct evaluation of the adequacy of specimens (i.e., whether they contain atypical cells), which could contribute to the pathological diagnosis of the target lesions. Therefore, the novel finding of the present study is that for small tumors (e.g., those of R-PC and BR-PC), especially when postoperative NTS is a concern owing to transgastric puncture of the body and tail, ROSE might provide a more accurate pathological diagnosis with EUS-TA with fewer punctures compared to those associated with SOSE (which allows indirect evaluation of the specimens). However, ROSE is not feasible in several medical facilities.23 We suggest that SOSE may serve as an alternative method for determining and minimizing the number of punctures based on the SVWC cutoff value.
In this study, the diagnosis rate showed no significant difference between specimens that did and did not meet the SVWC cutoff value. We attribute one reason for this lack of significance to the small number of samples that did not meet the SVWC cutoff value. Another reason is that the SVWC cutoff values employed in this study were set based on the results of previous studies in which the majority of patients had advanced cancer.21,22 Potentially, the SVWC cutoff value may need to be adjusted for smaller tumors, including R-PC and BR-PC. Nevertheless, the high sensitivity of the SVWC cutoff value for malignant diagnosis can serve as a reference for determining the number of punctures required for EUS-TA in medical facilities where ROSE is not feasible.
This study has some limitations. First, we included only two facilities; therefore, a multicenter study would have been ideal. Second, this study did not compare SOSE and ROSE; therefore, further prospective studies are required to determine the degree of additional effect of ROSE on the diagnostic results of SOSE.
In conclusion, EUS-TA with SOSE for small R-PC and BR-PC lesions had good, although somewhat insufficient, sensitivity for diagnosing malignancies based on the SVWC cutoff value. Developing an EUS-TA device and technique may be crucial to further improve our results.
Notes
Conflicts of Interest
The authors have no potential conflicts of interest.
Funding
None.
Author Contributions
Conceptualization: JI, KO, MW; Data curation: RH, TK, MT, TM, KA, TH, MK, CK; Formal analysis: HI, TI; Writing–original draft: JI, KO, MW; Writing–review & editing: all authors.