Gastrointestinal Endoscopy
Volume 69, Issue 6 , Pages 1103-1105, May 2009

PANDA cyst-fluid analysis: eats, shoots and leaves?

Department of Internal Medicine, Division of Gastroenterology, University of Michigan Health System, Ann Arbor, Michigan, USA

Article Outline

Abbreviations: CEA, carcinoembryonic antigen, PANDA, pancreatic-cyst–fluid DNA analysis

 

As the molecular evaluation of markers of malignant transformation have become commercialized, the allure of their potential utility has become irresistible.

Panda. Large black-and-white bear-like mammal, native to China. Eats, shoots and leaves.

In the popular book by Lynne Truss, Eats, Shoots & Leaves: The Zero Tolerance Approach to Punctuation,1 the author bemoans the state of punctuation in the United Kingdom and the United States and describes how rules are being relaxed in today's society. Her goal is to remind readers of the importance of punctuation in the English language by mixing humor and instruction. Truss dedicates the book “to the memory of the striking Bolshevik printers of St. Petersburg who, in 1905, demanded to be paid the same rate for punctuation marks as for letters, and thereby directly precipitated the first Russian Revolution.” The title of the book is an amphibology, a verbal fallacy arising from an ambiguous grammatical construction, and derived from a joke on bad punctuation:

“A panda walks into a café. He orders a sandwich, eats it, then draws a gun and proceeds to fire it at the other patrons. ‘Why?’ asks the confused, surviving waiter amidst the carnage, as the panda makes toward the exit. The panda produces a badly punctuated wildlife manual and tosses it over his shoulder. ‘Well, I'm a panda,’ he says, at the door. ‘Look it up.’ The waiter turns to the relevant entry in the manual and, sure enough, finds an explanation. “Panda. Large black-and-white bear-like mammal, native to China. Eats, shoots and leaves.”2

Medicine too, sometimes accepts new scientific information based upon ambiguous scientific data, just as we can be a bit sloppy with our grammar. The patient with a pancreatic cyst represents a great example of this clinical conundrum. Better imaging has found more lesions, and clinicians and patients alike struggle with how best to determine who should have a diagnostic or therapeutic intervention.3 As molecular evaluations of markers of malignant transformation have become commercialized, the allure of their potential utility has become irresistible.

The investigators of the PANDA (pancreatic-cyst–fluid DNA analysis) trial should be congratulated for their efforts in bridging molecular biology with clinical pancreatology and EUS. Their description of the strong association of mucinous cystic neoplasms with K-ras mutations that occur with other loss of heterozygosity mutations is a remarkable finding that elegantly correlates emerging data from in vitro and animal models with human research.4 Although results of previous studies suggest that mutations in the K-ras proto-oncogene are seen in hyperplastic foci from patients without neoplasia or chronic pancreatitis5, 6 and, therefore, nonspecific for pancreatic cancer, results of more recent research suggest that K-ras mutations do occur in association with dysplasia.7 Furthermore, a recent meta-analysis describes a stepwise increase in the frequency of K-ras mutations, with increasing degrees of dysplasia that support the idea that K-ras is an important early mediator of carcinogenesis in the pancreas.8 It is interesting to note that, in the study by Khalid et al,4 the presence of an isolated K-ras mutation was independently associated with the mucinous status in cystic lesions but did not differ significantly between malignant and premalignant mucinous cysts. This suggests that, although a K-ras mutation may be necessary for the formation of a malignant mucinous cyst of the pancreas, it is unlikely to be sufficient. This hypothesis is supported by mouse models in which constitutively expressed K-ras in combination with a Smad4/DPC4 deletion results in the formation of mucinous cystic lesions,9 but isolated K-ras mutations rarely lead to mucinous lesions or invasive phenotypes.10 Moreover, it was shown that nearly all invasive carcinomas arising in mucinous cystic neoplasms of the pancreas harbor mutations in the Smad4/DPC4 gene, which lead to a loss of expression.11 In this regard, the study by Khalid et al4 has done much to further our understanding of the mechanisms involved in cystic pancreatic tumorigenesis. But does their study erase the ambiguity in the clinical management of these patients and their cysts?

The investigators report that DNA analysis has better accuracy in diagnosing mucinous cysts and malignant cysts than current approaches (CEA [carcinoembryonic antigen] and cytology).4 However, the clinical design of the study is important to consider when evaluating the presented data. It is generally accepted that, when a new diagnostic test emerges, the test's performance should be compared with the currently accepted diagnostic test and that definitive information on the status of all patients (the proverbial “gold standard”) should be available to judge the true performance of the test in question.12, 13 In the current study, the investigators included only 113 of 299 patients and only those who underwent EUS-guided FNA and had definitive surgical pathology or malignant cytology. By excluding the other 186 patients, the investigators introduced an inherent selection bias, which potentially overestimated the performance of the DNA analysis and markedly restricted generalization of the results.

Currently, recommendations for surgical resection of mucinous cysts are based upon patient symptoms, EUS or other imaging results (including size, location, and appearance), and patient preference.14, 15 The clinical algorithm or decision process by which the investigators sent their patients to surgery was unfortunately not included in the paper. It would be ideal if the patients who underwent surgery were very similar to the larger population of 299 enrolled patients, but the data seem to suggest otherwise. The included patients had a higher percentage of malignant cysts among those with mucinous cysts and among those who had small (<3 cm) cysts than is reported in the literature.16, 17, 18 This, in turn, increases any pretest probability for malignant interpretation of the DNA analysis. Certainly, the included patients must have had one or more risk factors for malignant or mucinous cysts. In this unblinded study, it remains unclear whether or not the DNA analysis tipped the investigators to correctly triage patients to surgery. Because the multivariable analysis did not include EUS findings or patient symptoms in the predictive model for malignancy, might all of the malignant lesions been correctly sent to surgery without DNA testing anyway? Indeed, the presence of a solid component in the cyst was 75% sensitive for the detection of a malignant lesion and 84% specific for the detection of either a malignant lesion or a mucinous lesion, approaching the performance of the far more costly DNA analysis. Moreover, without including the test performance in the excluded patients, this study does not inform us what each of the analyses predicts in patients with low-risk EUS findings (such as a simple small cyst ≤3 cm) and no symptoms.

Given that the DNA analysis is significantly more expensive than both cyst-fluid CEA and cytology, the ultimate question faced by clinicians is whether DNA analysis of cyst fluid provides additional clinically meaningful information to justify the effort and cost of the test. The PANDA study gives us some hints that the answer may be yes but likely in very highly selected circumstances. What is clear is that DNA analysis can point to a mucinous lesion when there is some uncertainty from the CEA analysis alone. The DNA testing requirement for a small volume may make this differentiation easier in some centers, because CEA levels were available in only 2 of 3 patients in this trial (which seems quite low compared with our clinical experience). In some patients with large lesions (particularly those having negative cytology and low CEA levels) in which differentiation from a pseudocyst may be difficult or clinical suspicion for a mucinous lesion is high, we envision a role for this analysis. For small lesions likely to undergo surveillance anyway, it would be difficult to justify the cost and effort, particularly in light of growing evidence that suggests that lesions smaller than 3 cm in size are unlikely to grow or develop malignant transformation.19, 20 If future longitudinal studies show that some of these tiny cysts with benign features on EUS carry a clinically important risk of malignancy, then the role of DNA analysis could be expanded.

Our own experience is that the risk of malignancy in cystic lesions resected without worrisome features and without symptoms is ≤4%.21 In contrast to the experience of most tertiary-care centers,18, 19, 22 the PANDA study reports 25 out of 40 (nearly 63%) of their malignant cysts were ≤3 cm in size. This suggests that there were other markers that clearly pointed to malignancy in these patients or that significant weight was given to the results of the DNA analysis. Although clinicians were not blinded to the DNA analysis results, it would have been informative to know how many of these malignancies would have been subjected to surgery if the DNA results were not known.

The investigators conclude that DNA analysis should be considered when cyst cytology is negative for malignancy. However, their data simply do not support this recommendation except in highly selected circumstances. Further, because no DNA analysis data are provided on those who did not have surgery (all of whom had negative cytology), the key issue of how often DNA analysis correctly points to the need for surgery (malignant or advanced premalignant lesion) has not been answered. A number of recent abstracts reported that worrisome DNA analyses obtained in routine clinical practice led to surgery in patients with benign lesions, such as those with pseudocysts.

Should the current approach, which combines the recommendations of the consensus document23 and other experts, be altered by the results of this study? We propose that a study involving the long-term follow-up of patients who are not subjected to surgery but who have DNA analyses will be the only way the true incremental value of this test will be revealed. When the true performance characteristics of this approach are properly defined, we can appropriately address the issue of incremental cost-effectiveness, as well as clinical utility. For now, published treatment algorithms have provided excellent clinical outcomes without incorporation of additional DNA analyses.22

So, returning to the PANDA of our title, should we EUS, FNA for CEA and cytology and leave, or add DNA analysis as well? We conclude—not yet. But we remain excited at the potential for new biomarkers in this field, in which the impact on clinical decision making can be expected to lead to markedly improved patient outcomes.

Back to Article Outline

Disclosure 

All authors disclosed no financial relationships relevant to this publication.

Back to Article Outline

References 

  1. Truss L. Eats, shoots and leaves. New York: Gotham Books; 2004;
  2. Eats, shoots and leaves. Wikipedia 2008. Available at: http://en.wikipedia.org/wiki/Eats,_Shoots_&_Leaves. Accessed September 2, 2008.
  3. Scheiman JM. Cystic lesion of the pancreas. Gastroenterology. 2005;128:463–469
  4. 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
  5. Berger DH, Chang H, Wood M, et al. Mutational activation of K-ras in nonneoplastic exocrine pancreatic lesions in relation to cigarette smoking status. Cancer. 1999;85:326–332
  6. Tada M, Ohashi M, Shiratori Y, et al. Analysis of K-ras gene mutation in hyperplastic duct cells of the pancreas without pancreatic disease. Gastroenterology. 1996;110:227–231
  7. Jeong S, Lee DH, Lee JI, et al. Expression of Ki-67, p53, and K-ras in chronic pancreatitis and pancreatic ductal adenocarcinoma. World J Gastroenterol. 2005;11:6765–6769
  8. Löhr M, Klöppel G, Maisonneuve P, et al. Frequency of K-ras mutations in pancreatic intraductal neoplasias associated with pancreatic ductal adenocarcinoma and chronic pancreatitis: a meta-analysis. Neoplasia. 2005;7:17–23
  9. Izeradjene K, Combs C, Best M, et al. Kras(G12D) and Smad4/Dpc4 haploinsufficiency cooperate to induce mucinous cystic neoplasms and invasive adenocarcinoma of the pancreas. Cancer Cell. 2007;11:229–243
  10. Hingorani SR, Petricoin EF, Maitra A, et al. Preinvasive and invasive ductal pancreatic cancer and its early detection in the mouse. Cancer Cell. 2003;4:437–450
  11. Iacobuzio-Donahue CA, Wilentz RE, Argani P, et al. Dpc4 protein in mucinous cystic neoplasms of the pancreas: frequent loss of expression in invasive carcinomas suggests a role in genetic progression. Am J Surg Pathol. 2000;24:1544–1548
  12. Jaeschke R, Guyatt G, Sackett DL. Users' guides to the medical literature. III. How to use an article about a diagnostic test. A. Are the results of the study valid? Evidence-Based Medicine Working Group. JAMA. 1994;271:389–391
  13. Jaeschke R, Guyatt GH, Sackett DL. Users' guides to the medical literature. III. How to use an article about a diagnostic test. B. What are the results and will they help me in caring for my patients? The Evidence-Based Medicine Working Group. JAMA. 1994;271:703–707
  14. Scheiman JM. Management of cystic lesions of the pancreas. J Gastrointest Surg. 2008;12:405–407
  15. Brugge WR, Lauwers GY, Sahani D, et al. Cystic neoplasms of the pancreas. N Engl J Med. 2004;351:1218–1226
  16. Brugge WR, Lewandrowski K, Lee-Lewandrowski E, et al. Diagnosis of pancreatic cystic neoplasms: a report of the cooperative pancreatic cyst study. Gastroenterology. 2004;126:1330–1336
  17. Sahani DV, Saokar A, Hahn PF, et al. Pancreatic cysts 3 cm or smaller: how aggressive should treatment be?. Radiology. 2006;238:912–919
  18. Crippa S, Salvia R, Warshaw AL, et al. Mucinous cystic neoplasm of the pancreas is not an aggressive entity: lessons from 163 resected patients. Ann Surg. 2008;247:571–579
  19. Rodriguez JR, Salvia R, Crippa S, et al. Branch-duct intraductal papillary mucinous neoplasms: observations in 145 patients who underwent resection. Gastroenterology. 2007;133:72–79quiz 309-10
  20. Allen PJ, D'Angelica M, Gonen M, et al. A selective approach to the resection of cystic lesions of the pancreas: results from 539 consecutive patients. Ann Surg. 2006;244:572–582
  21. Lee CJ, Scheiman J, Anderson MA, et al. Risk of malignancy in resected cystic tumors of the pancreas < or =3 cm in size: is it safe to observe asymptomatic patients? A multi-institutional report. J Gastrointest Surg. 2008;12:234–242
  22. Tang RS, Weinberg B, Dawson DW, et al. Evaluation of the guidelines for management of pancreatic branch-duct intraductal papillary mucinous neoplasm. Clin Gastroenterol Hepatol. 2008;6:815–819
  23. Tanaka M, Chari S, Adsay V, et al. International consensus guidelines for management of intraductal papillary mucinous neoplasms and mucinous cystic neoplasms of the pancreas. Pancreatology. 2006;6:17–32

PII: S0016-5107(08)02824-1

doi:10.1016/j.gie.2008.10.034

Gastrointestinal Endoscopy
Volume 69, Issue 6 , Pages 1103-1105, May 2009

Article Tools