Reply to Barresi et al.

We thank Dr. Barresi and his colleagues for their interest in our work. Between-study comparisons of performance characteristics are valid only when the clinical end points and patient populations examined are comparable. These features differed significantly between our study [1] and the study of Goh et al. [2], accounting for the difference in positive predictive values (PPVs).

The PPV (88 %) reported by Goh et al. for the “modified” Sendai 2012 guidelines was related to their performance in detecting “malignant or potentially malignant” pancreatic cysts, which included all main-duct/mixed-type intraductal papillary mucinous neoplasms (MD/MT-IPMNs) and mucinous cystic neoplasms (MCNs), regardless of outcome for benign or malignant disease, whereas our PPV (21 %) was related to the performance of the Sendai 2012 guidelines in detecting malignant or benign cysts based on clinical outcomes. Additionally, the prevalence of malignancy strongly impacts the PPV of any diagnostic criterion or test, and the prevalence of malignancy differed between the two study populations. Goh et al. examined a surgery-only population. As stated by the authors, this population was therefore enriched for malignant cases, with a diminished number of benign cases; thus, a reduced number of possible false-positive cases relative to the number of true-positive cases provided a seemingly higher PPV. The prevalence of malignancy in our study population (patients followed clinically and patients undergoing surgery) was approximately 10 %; this prevalence better represents the real-world population but remains somewhat higher than the true prevalence of malignancy. Therefore, in clinical practice, the PPV for the Sendai 2012 guidelines, and for integrated molecular pathology (IMP), would be expected to be lower. Regardless, adjunctive IMP testing adds value to the Sendai 2012 guidelines by reducing the number of false-positive results without missing more malignancies.

The lower PPV of the Sendai 2012 guidelines in our study has been substantiated in independent studies assessing their performance in detecting malignancy based on patient outcomes: 22.5 % in those undergoing endoscopic ultrasound (EUS) for all pancreatic cysts and 29 % to 30 % in those undergoing surgery for branch-duct intraductal papillary mucinous neoplasms (BD-IPMNs) based on worrisome features [3] [4]. The prevalences of malignancy and the clinical end points of benign and malignant disease in these studies were similar to those in our study.

As noted in our report, we considered cyst size larger than 3 cm an indication for surgery but recognize that the Sendai 2012 guidelines regarding size are slightly more complex. The Sendai 2012 guidelines recommend resection for all surgically fit patients with MCNs; thus, we included cytologic evidence of mucin as a criterion for surgery, but not a carcinoembryonic antigen (CEA) level of 192 ng/mL or higher (because this lowered the PPV to 18 % in a separate analysis). However, as reported previously (based on the relative risk summary statistic owing to constraints on manuscript length), omitting these indications completely had minimal impact on the Sendai PPV: 24 % when cyst size was excluded and 26 % when the presence of mucin was excluded, versus 21 % when both were included.

The four molecular criteria reported are those used as standard in the commercial IMP test; the report of Khalid et al. [5] was cited to provide supportive evidence for these criteria but was not intended to define them completely.

In summary, the PPV for the Sendai 2012 guidelines in our study and in the study of Goh et al. cannot be compared because of the differences in the clinical end points assessed and in the prevalence rates of malignancy in the patient populations examined. However, our findings are supported by other studies assessing comparable end points and populations.

• References

• 1 Al-Haddad MA, Kowalski T, Siddiqui A et al. Integrated molecular pathology accurately determines the malignant potential of pancreatic cysts. Endoscopy 14.10.2014; DOI: DOI: 10.1055/s-0034-1390742. [Epub ahead of print].
• 2 Goh BK, Tan DM, Thng CH et al. Are the Sendai and Fukuoka consensus guidelines for cystic mucinous neoplasms of the pancreas useful in the initial triage of all suspected pancreatic cystic neoplasms? A single-institution experience with 317 surgically-treated patients. Ann Surg Oncol 2014; 21: 1919-1926
• 3 Lawson R, Savides T, Kwong W et al. Natural history of pancreatic cystic lesions: validation of the 2012 Sendai criteria. Am J Gastroenterol 2013; 108: S91 Abstract 301
• 4 Sahora K, Mino-Kenudson M, Brugge W et al. Branch duct intraductal papillary mucinous neoplasms: does cyst size change the tip of the scale? A critical analysis of the revised international consensus guidelines in a large single-institutional series. Ann Surg 2013; 258: 466-475
• 5 Khalid A, Zahid M, Finkelstein SD et al. Pancreatic cyst fluid DNA analysis in evaluating pancreatic cysts: a report of the PANDA study. Gastrointest Endosc 2009; 69: 1095-1102