Suspicious Cytologic Diagnostic Category in Endoscopic Ultrasound-Guided FNA of the Pancreas: Follow-Up and Outcomes Evan A. Alston, MD 1 ; Sejong Bae, PhD 2 ; and Isam A. Eltoum, MD, MBA 1 BACKGROUND: The objective of the current study was to assess how the suspicious category is followed up in a large endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA) service, its outcomes, and the predictors that are likely to be associated with the subsequent diagnosis of a neoplastic process. METHODS: For pancreatic EUS-FNA samples with the suspicious category diagnosis, the authors reviewed the electronic medical record for the method of follow-up and the risks associated with pancreatic malignancy. Logistic regression analysis was used to determine the risk factors that were likely to be associated with the diagnosis of a neoplastic lesion after a cytologic diagnosis of suspicious. RESULTS: Of a total of 3832 EUS-FNA cases, 116 were diagnosed with suspicious cytology. A total of 90 of 98 neoplasms (92%) were identified, including 72 carcinomas (73%). Similar rates of neoplasia were detected after repeat FNA (34 of 37 neoplasms [92%]) and subsequent biopsy/surgical resection (44 of 46 neoplasms [96%]), but significantly fewer neoplasms were detected among patients with clinical follow-up (18 of 23 neoplasms [78%]). On multivariate analysis of the potential predictive variables listed above, the presence of a mass was found to be significantly associated with a higher rate of diagnosis of a neoplasm, whereas weight loss was significantly associated with a diagnosis of carcinoma. CONCLU- SIONS: The diagnostic category of suspicious is associated with a high risk of benign and malignant neoplasms, regardless of the method of follow-up. The presence of a mass and weight loss are significant predictors of a subsequent diagnosis of a neoplasm after suspicious cytology. In patients with suspicious cytology and these findings, surgery is recommended for resectable masses and repeat FNA for unresectable masses. Cancer (Cancer Cytopathol) 2016;124:53-7. VC 2015 American Cancer Society. KEY WORDS: diagnostic accuracy; endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA); guidelines; pancreas; suspicious. INTRODUCTION The development and standardization of endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA) of the pancreas has provided a reliable and minimally invasive means of evaluating clinically suspicious lesions. 1 4 Although many cytopathologists follow Bethesda System-like categories (unsatisfactory, benign, atypical, suspicious, and positive for neoplasm), the terminology and specific criteria for cytologic diagnosis in this organ have varied among institutions. 5 Recently, the Papanicolaou Society of Cytopathology published guidelines for the interpretation and follow-up of pancreatobiliary cytology, 6 including the categories of nondiagnostic, negative, atypical, suspicious, and positive/malignant, as well as neoplastic (benign or other). Whereas more rigorously defined categories will clarify pathologic interpretations for clinicians, the indeterminate categories of atypical and suspicious continue to be, as in most organs, the most difficult for clinicians to interpret and manage. The Corresponding author: Evan A. Alston, MD, Division of Anatomic Pathology, Department of Pathology, University of Alabama at Birmingham, 619 South 19th St, Birmingham, AL 35233; Fax: (205) 975-5242; eaalston@uabmc.edu 1 Division of Anatomic Pathology, Department of Pathology, University of Alabama at Birmingham, Birmingham, Alabama; 2 Division of Preventive Medicine, Department of Medicine, University of Alabama at Birmingham, Birmingham, Alabama Received: June 1, 2015; Revised: July 23, 2015; Accepted: August 5, 2015 Published online September 3, 2015 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/cncy.21611, wileyonlinelibrary.com Cancer Cytopathology January 2016 53
objective of the current study was to assess the types of follow-up provided for cases diagnosed as suspicious in a large EUS-FNA service, the outcomes of these cases, and the possible predictors of neoplastic/malignant outcomes. MATERIALS AND METHODS In the current retrospective study, we collected information regarding all cases diagnosed as suspicious by EUS- FNA from July 2000 through March 2013 at the University of Alabama at Birmingham. Electronic records were reviewed for follow-up and for the risks associated with pancreatic neoplasia. The final outcome was based on surgical diagnosis, repeat FNA, or clinical follow-up. Cases were considered positive for neoplasm if a neoplasm was demonstrated on repeat FNA; on histology; by progression of the disease with clinical follow-up; or by the diagnosis of metastatic disease through imaging, cytology, or histologic examination. Data collected from the chart review included the diagnosis; patient age; patient sex; time to last follow-up; and information regarding smoking history, alcohol abuse, diabetes, jaundice, abdominal pain, history of pancreatitis, weight loss, the presence of a mass, obstruction of the bile duct, or dilation of the pancreatic duct. EUS-FNA was performed under conscious sedation using a curvilinear echoendoscope (UC-30P; Olympus America, Melville, NY) as described previously. 7 FNAs were performed using a 22-gauge needle, with 19-gauge or 25-gauge needles used in a few cases. A cytopathologist was present for the onsite evaluation of adequacy and the diagnosis of all pancreatic masses during EUS-FNA. For onsite examinations, only Diff-Quik staining (Baxter Healthcare, Deerfield, Ill) was used. For final pathological interpretation, additional Papanicolaou-stained slides were prepared, as well as hematoxylin and eosin stains for cell blocks. During the study period, multiple pathologists and clinicians were involved in the diagnostic process and the clinical decision-making regarding the management of these patients. We used the following diagnostic categories: positive for neoplasm, suspicious for neoplasm, atypical with a comment, benign, and unsatisfactory. Cell blocks were routinely ordered and examined for further analysis. Flow cytometry was performed for atypical lymphoid infiltrates. The management of and decision regarding the type of follow-up of a case after the diagnosis was largely based on clinical, imaging, and cytologic findings as we have described previously. 8 At the study institution, there were no defined criteria for the suspicious category, but in general these specimens either had cytologic atypia severe enough that the pathologist suspected that a ductal adenocarcinoma was present but lesional cells were insufficient in quantity to make a diagnosis of positive for malignancy, or there were characteristic features of a neoplasm other than ductal adenocarcinoma (including lymphoma, pancreatic neuroendocrine tumor, intraductal papillary mucinous neoplasm, solid pseudopapillary neoplasm, and acinar cell carcinoma) but insufficient material for confirmation with ancillary studies. The atypical category similarly was based on the pathologists judgment that the level of atypia was more than benign or reactive but insufficient for a diagnosis of suspicious. Although there were no formal consensus meetings held in the study department to define the suspicious category before this series, intradepartmental consultations were used frequently for indeterminate diagnoses. Statistical Analysis Statistical analysis was performed using SPSS statistical software (version 21.0; IBM Corporation, Armonk, NY). The main analysis focused on the presence of a neoplasm as an endpoint. In some analyses, particularly as related to risk, the presence of ductal carcinoma was considered to be a secondary outcome. The suspicious cases were crosstabulated by the method of follow-up, and the rates of neoplasm detection were then compared among the different methods using the chi-square test. RESULTS Of 3832 EUS-FNA cases, 116 (3%) (58% of which were from men with a mean age of 66 years [standard deviation, 12 years]) were diagnosed with suspicious cytology. The overall median follow-up was 131 days, and 18 patients (16%) were lost to follow-up. Follow-up included 29 repeat FNAs (25%), 38 subsequent biopsies/ surgical resections (33%), 8 cases of both repeat FNA and biopsy/surgical resection (7%), and 23 cases with clinical follow-up (20%). After a cytologic diagnosis of suspicious, a total of 90 of 98 neoplasms (92%) were identified, including 72 ductal carcinomas (73%). Similar rates of neoplasia were detected after repeat FNA (34 of 54 Cancer Cytopathology January 2016
Suspicious EUS-FNA of the Pancreas/Alston et al TABLE 1. Overall Frequency of Predictive Factors of Neoplasm Factor Frequency (%) Mass present 92 (93.9) Alcohol abuse 17 (17.3) Male gender 58 (59.2) Jaundice 46 (46.9) Abdominal pain 61 (62.2) Weight loss 61 (62.2) Dilation of pancreatic duct 43 (43.9) Obstruction of bile duct 28 (28.6) Smoking history 49 (50) Diabetes 25 (25.5) History of pancreatitis 7 (8.16) 37 neoplasms [92%]) and subsequent biopsy/surgical resection (44 of 46 neoplasms [97%]), but significantly fewer neoplasms were detected among patients with clinical follow-up (18 of 23 neoplasms [78%]). There were 8 cases initially diagnosed as suspicious with no neoplasm identified on follow-up, which included subsequent diagnoses of pancreatitis (2 cases), fibrosis and chronic inflammation suggestive of autoimmune pancreatitis (1 case), and no disease progression with documented clinical follow-up (5 cases). There were also 18 neoplasms identified on follow-up that were not ductal adenocarcinoma, and these included large B-cell lymphoma (8 neoplasms), neuroendocrine neoplasm (6 neoplasms), intraductal papillary mucinous neoplasm (3 neoplasms), and solid pseudopapillary neoplasm (1 neoplasm). These neoplasms were diagnosed based on ancillary studies of subsequent specimens, which could not be performed on the initial EUS-FNA specimens due to inadequate cellularity. The collection of signs and symptoms suggestive of a pancreatic neoplasm on initial clinical evaluation varied among patients with a cytologic diagnosis of suspicious (Table 1). Of the 98 patients who had follow-up, 92 (94%) had a definitive mass, 49 patients (50%) were smokers, 17 patients (17%) consumed alcohol, 25 patients (26%) were diabetic, 46 patients (47%) were jaundiced, 61 patients (62%) had abdominal pain, 61 patients (62%) had weight loss, 8 patients (8%) had a history of pancreatitis, 28 patients (29%) had an obstructed bile duct, and 43 patients (44%) had a dilated pancreatic duct. On multivariate analysis of the potential predictive variables listed above, the presence of a mass was found to be significantly associated with a higher rate of diagnosis of a neoplasm (Table 2), whereas weight loss was found to be significantly associated with ductal carcinoma (Table 3). TABLE 2. Multivariate Analysis of Predictive Factors of Neoplasm Factor b Coefficient (SE) P Smoking history 20.38 (0.96).696 Alcohol abuse 0.87 (1.22).476 Male sex 0.09 (59.2).920 Age 20.01 (0.04).738 Jaundice 0.58 (0.98).554 Abdominal pain 20.76 (0.96).431 Weight loss 20.25 (0.97).798 Obstruction of bile duct 21.25 (1.20).300 Dilation of pancreatic duct 20.19 (0.90).831 Mass present 2.62 (1.19).027 Abbreviation: SE, standard error. TABLE 3. Multivariate Analysis of Predictive Factors of Ductal Carcinoma Factor b Coefficient (SE) P Smoking history 0.27 (0.58).641 Alcohol abuse 0.01 (0.78).994 Jaundice 20.48 (0.60).424 Weight loss 21.20 (0.55).029 Obstruction of bile duct 21.17 (0.81).148 Dilation of pancreatic duct 20.65 (0.58).262 Age 0.03 (0.02).183 Abbreviation: SE, standard error. DISCUSSION As might be expected, the suspicious for neoplasm category is highly associated with the presence of either a benign or malignant neoplasm, but despite this association, the method of follow-up for the suspicious category is variable and a substantial percentage of cases in the current series (16%) were lost to follow-up. The only variable found to be associated with the presence of a neoplasm after a suspicious cytologic diagnosis was the presence of a mass on imaging. The vast majority of cases with suspicious cytology had a mass noted on imaging (94%), but this variable was not found to be significantly associated with the presence of ductal carcinoma, thereby limiting its usefulness in making decisions concerning follow-up. In the current study, the only variable found to be significantly associated with the presence of ductal carcinoma after a suspicious diagnosis was weight loss. Although this clinical finding may be reason to suggest surgical intervention as opposed to clinical follow-up, its significance should be confirmed by additional studies performed at other institutions. Regardless, the high rate of ductal carcinoma in all neoplasms detected in the current study (73%) underscores the importance of close follow-up for this Cancer Cytopathology January 2016 55
diagnostic category. In the study institution, the majority of cases were followed with either repeat FNA, surgical biopsy, or both (64%). However, sizeable percentages of cases had only clinical follow-up (20%) or were lost to follow-up (16%). For many of the patients with clinical follow-up alone, there was a strong clinical suspicion of malignancy before the FNA procedure, and many of these patients opted for hospice care without further intervention. Many of the patients who were lost to follow-up were transferred to outside facilities, presumably for additional follow-up or palliative care. Although management decisions depend on a particular patient s comorbidities and preferences, the challenge for clinicians lies in determining the appropriate follow-up method for suspicious lesions, given their own pretest clinical suspicion. As we have previously suggested, 5 a predictive model that takes into account clinical signs and symptoms of pancreatic malignancy could potentially improve the overall usefulness of EUS-FNA cytology by further characterizing an individual patient s risk of malignancy after an indeterminate cytologic diagnosis. The incorporation of clinical and radiologic information with cytologic diagnoses for risk stratification, akin to the triple test used for breast lesions, 9 would require further studies including data from multiple institutions, but potentially could provide a more reliable and evidence-based approach to the difficult treatment decisions regarding patients with indeterminate pancreatic lesions. There are limitations to the current study. Although it is based on a retrospective review of a large number of suspicious cases over a long time period, these cases were taken from a single institution. Although there was a relatively long median follow-up period (131 days), a large percentage of cases (16%) were lost to clinical follow-up and it is unclear whether there was a low or high clinical suspicion of malignancy in these cases. The high number of patients without follow-up likely affected the observed rate of malignancy in the current series, given the clinical selection bias for follow-up. It is likely that the methods of clinical follow-up and interpretation of the suspicious cytologic category vary between institutions and additional studies that compile cases from different institutions are needed. A recent analysis by Layfield et al 10 of pancreatic EUS-FNA diagnoses at 4 large medical centers documented higher rates of malignancy after atypical (79%) and suspicious (96%) diagnoses compared with our series, 11 and the authors suggested that although the suspicious category is statistically distinct, it is not clear whether it is clinically distinct from the atypical category. Although we believe that the suspicious diagnosis category is useful as a category distinct from atypical diagnoses, we agree that the diagnostic criteria for atypical and suspicious diagnoses should correspond to clinically meaningful decision points. The recently published Papanicolaou Society of Cytopathology guidelines for pancreatobiliary cytology offer a more rigorous set of criteria for the indeterminate categories than was available when this series of cases was originally diagnosed, and agreement among cytopathologists regarding the severity of atypia present in the indeterminate diagnoses certainly will clarify the meaning of these categories. We expect that additional studies that assess the outcomes of indeterminate lesions will provide evidence for the appropriate follow-up in these cases, potentially in the form of a predictive model of clinical risk. Given the high rate of malignancy that we documented in our series of suspicious cases, we currently believe that a conservative follow-up approach is necessary, with surgical resection performed after the initial suspicious FNA if clinically appropriate. To our knowledge, the current study is the largest study of suspicious diagnoses by EUS-FNA reported to date. It demonstrates the variability in methods of followup for this diagnosis, despite the high rate of neoplasms detected. Although the method of follow-up largely depends on the pretest clinical suspicion of the lesion and patient comorbidities, the only clinical factor found to be significantly associated with ductal carcinoma was a history of weight loss. Although patient characteristics and comorbidities must be taken into account, we recommend that any resectable lesion with a suspicious diagnosis be followed with surgical biopsy, and any unresectable lesion be followed with repeat FNA. FUNDING SUPPORT No specific funding was disclosed. CONFLICT OF INTEREST DISCLOSURES Dr. Eltoum was supported by a grant from the National Institutes of Health for work performed as part of the current study. REFERENCES 1. Hebert-Magee S, Bae S, Varadarajulu S, et al. The presence of a cytopathologist increases the diagnostic accuracy of endoscopic 56 Cancer Cytopathology January 2016
Suspicious EUS-FNA of the Pancreas/Alston et al ultrasound-guided fine needle aspiration cytology for pancreatic adenocarcinoma: a meta-analysis. Cytopathology. 2013;24:159 171. 2. Puli SR, Bechtold ML, Buxbaum JL, Eloubeidi MA. How good is endoscopic ultrasound-guided fine-needle aspiration in diagnosing the correct etiology for a solid pancreatic mass?: a meta-analysis and systematic review. Pancreas. 2013;42:20 26. 3. Affolter KE, Schmidt RL, Matynia AP, Adler DG, Factor RE. Needle size has only a limited effect on outcomes in EUS-guided fine needle aspiration: a systematic review and meta-analysis. Dig Dis Sci. 2013;58:1026 1034. 4. Hewitt MJ, McPhail MJ, Possamai L, Dhar A, Vlavianos P, Monahan KJ. EUS-guided FNA for diagnosis of solid pancreatic neoplasms: a meta-analysis. Gastrointest Endosc. 2012;75:319 331. 5. Abdelgawwad MS, Alston E, Eltoum IA. The frequency and cancer risk associated with the atypical cytologic diagnostic category in endoscopic ultrasound-guided fine-needle aspiration specimens of solid pancreatic lesions: a meta-analysis and argument for a Bethesda System for Reporting Cytopathology of the Pancreas. Cancer (Cancer Cytopathol). 2013;121:620 628. 6. Pitman MB, Centeno BA, Ali SZ, et al; Papanicolaou Society of Cytopathology. Standardized terminology and nomenclature for pancreatobiliary cytology: the Papanicolaou Society for Cytopathology guidelines. Diagn Cytopathol. 2014;42:338 350. 7. Eloubeidi MA, Tamhane A. Prospective assessment of diagnostic utility and complications of endoscopic ultrasound-guided fine needle aspiration. Results from a newly developed academic endoscopic ultrasound program. Dig Dis. 2008;26:356 363. 8. Eloubeidi MA, Varadarajulu S, Desai S, et al. A prospective evaluation of an algorithm incorporating routine preoperative endoscopic ultrasound-guided fine needle aspiration in suspected pancreatic cancer. J Gastrointest Surg. 2007;11:813 819. 9. The uniform approach to breast fine-needle aspiration biopsy. National Cancer Institute Fine-Needle Aspiration of Breast Workshop Subcommittees. Diagn Cytopathol. 1997;16:295 311. 10. Layfield LJ, Schmidt RL, Hirschowitz SL, Olson MT, Ali SZ, Dodd LL. Significance of the diagnostic categories atypical and suspicious for malignancy in the cytologic diagnosis of solid pancreatic masses. Diagn Cytopathol. 2014;42:292 296. 11. Alston E, Bae S, Eltoum IA. Atypical cytologic diagnostic category in EUS-FNA of the pancreas: follow-up, outcomes, and predictive models. Cancer (Cancer Cytopathol). 2014;122:428 434. Cancer Cytopathology January 2016 57