Novitas LCD January 2025 PANCRAGEN

 Novitas LCD L39367

LCD:

https://www.cms.gov/medicare-coverage-database/view/lcd.aspx?lcdid=39367&ver=28&contractorName=all&contractorNumber=all&updatePeriod=1129&sortBy=id&bc=13

January 27, 2025 blog to which this page is a sidebar:

https://www.discoveriesinhealthpolicy.com/2025/01/genomeweb-scoop-novitas-delays-cancer.html

SIDEBAR:  PANCRAGEN IN JANUARY 2025 NOVITAS LCD L39367

PancraGEN- Interpace Diagnostics - DATA REVIEW

In addition to the articles submitted with comments, PubMed and Google Scholar were searched for peer-reviewed, evidence-based literature that provided information regarding the analytic and clinical validity and clinical utility for the PancraGEN test. Key words used to search in combination included: PancraGEN, PathfinderTG, molecular testing, topographic genotyping, pancreatic cyst(s), pancreatic cyst fluid, solid pancreatic lesions, and KRAS and/or GNAS mutations.

Thirty-five total publications addressing the analytical validity, clinical validity, and/or clinical utility of the PancraGEN prognostic test were identified. The papers identified focused on both individuals with pancreatic cysts and with solid pancreaticobiliary lesions.

In 2006, a patent was filed for a topographic genotyping molecular analysis test, which would later become PathfinderTG and then PancraGEN. The test was designed to classify the risk of pancreatic cysts and solid pancreaticobiliary lesions when first line evaluation results were inconclusive.54 According to the patent, the test directly measured several aspects of a specimen: DNA quality, loss of heterozygosity (LOH) in tumor suppressor genes, mutations in oncogenes (only K-ras oncogene specifically named), other less well defined genetic targets (e.g., “structural alterations in DNA”), percentage of mutated DNA per identified DNA abnormality, and “specific temporal sequence of mutation accumulation” as determined from the aforementioned percentages of mutated DNA. Altogether, these measurements would be used for “diagnosing and/or determining the prognosis of a pancreatic anomaly in a patient suffering from pancreatic cysts” and used for “determining a course of treatment for a pancreatic anomaly.”

Since the original patent, significant changes have been made to the original test’s data input and the presentation of test results. In 2012, the test results were changed from 3 categorical results (benign, statistically indolent, and aggressive) to 4 results (benign, statistically indolent, statistically higher-risk, and aggressive).55 Since result categories are tied to specific prognostic outcomes and advise different next steps in clinical care, changing the number and type of categories changes the test output, thereby creating a new test. The most recent version of PancraGEN added analysis of the GNAS oncogene. Since mutations in GNAS would be considered a “significant molecular alteration,” testing of GNAS would potentially reclassify any specimens that had been classified based on KRAS alone. Considering the result categories and genes analyzed were both changed, the latest version of PancraGEN is a new, distinct test. The literature for older versions of PathfinderTG are not comparable to the current version of PancraGEN.

Of the 35 publications identified, 26 publications described an earlier version of the PancraGEN test that utilized less molecular data and provided fewer categorical results than the currently offered PancraGEN test.56-81 The Agency for Healthcare Research and Quality (AHRQ) performed a technical review of an earlier version of PancraGEN.80 The 2015 AGA Guidelines for diagnosis and management of pancreatic cysts does not name PancraGEN directly in its section about molecular testing, but instead only cites 2 papers discussing older versions of the PancraGEN/Pathfinder test.73

Five of the 35 publications identified did not analyze the PancraGEN test’s primary output (categorical results) but instead evaluated specific components of the test (i.e., molecular test data).82-86. In fact, the study by Shen and colleagues stated that it was not meant “to evaluate the scientific methods or validity of this commercially available test.” 85

Of the 35 publications identified, there were only 4 papers that evaluated the current version of the PancraGEN including the 4 categorical results. One retrospective study addressed PancraGEN’s clinical validity and clinical utility for pancreatic cysts.55 Two retrospective studies utilizing data from Al-Haddad and colleagues’ study addressed the clinical utility of PancraGEN for pancreatic cysts.87,88 One paper addressed PancraGEN’s clinical validity and clinical utility for solid pancreaticobiliary lesions.89

###

PancraGEN- Interpace Diagnostics - CONCLUSIONS

PancraGEN (also known as Pathfinder TG and Integrated molecular pathology [IMP]) has received multiple updates to its input data and algorithmic categorization of risk since its initial release. Comparison of early example reports to the most recent example report (available on the PancraGEN website) clearly demonstrate this evolution.136,137 The current version of the PancraGEN report relies on algorithmic assessment of molecular data, cyst fluid test results, and radiologic findings to determine a patient’s risk for developing high grade dysplasia (HGD) and/or carcinoma. The algorithm is clearly described and diagrammed in the sample report.137 The algorithmic stratification of risk is heavily weighted towards molecular data, with the absence of “significant molecular alterations” automatically resulting in “Benign” categorization and the presence of 2 or more “significant molecular alterations” automatically resulting in “Aggressive” categorization. The “Benign” category confers a “97% probability of benign disease over the next 3 years” and the “Aggressive” category confers a “91% probability of HGD/carcinoma”. Per the sample report, 5 “significant molecular alterations” are described137:

1. 1. “High levels of DNA”
   2. “High clonality KRAS point mutation”
   3. “High clonality GNAS point mutation”
   4. “Single high clonality LOH tumor suppressor gene mutation”
   5. “Two or more low clonality LOW tumor suppressor gene mutations”

The current body of literature does not support the clinical validity of the PancraGEN test. Since the algorithm primarily categorizes risk via “significant molecular alterations” and the “Benign” category is defined as absence of these alterations, not testing any of the above 5 alterations would result in an underestimation of patients with potential higher risk of HGD/carcinoma. Therefore, adequate assessment of clinical validity of the current version of PancraGEN would require assessment of all 5 alterations in study populations. None of the 4 studies assessing the current version of PancraGEN fully assess all 5 alterations.55,87-89 For example, in the 3 retrospective studies derived from National Pancreatic Cyst Registry, a significant number of patients (“468/492 IMP diagnoses”) were NOT tested for GNAS because their data was collected from earlier versions of the PancraGEN test that did not include GNAS testing.55,87,88 Another study, from Khosravi and colleagues, addresses the 4 categorical results (simplifying them into 2 categories for the paper: low and high risk) but does not discus GNAS or clonality of identified mutations. As a result, PancraGEN studies lack the statistical integrity required to establish the clinical validity of the current version of PancraGEN.

Additionally, there are no studies supporting the clinical utility of the PancraGEN test.

First, there are no prospective studies for PancraGEN that directly evaluate its effect on patient management and outcome. The 4 studies evaluating the current version of PancraGEN are all retrospective, assessing patient populations who received PancraGEN testing as part of their clinical care; however, the assessment of PancraGEN’s effect on patient management and outcome is extrapolated from reading patient charts after the fact.

Second, cysts with a potential to develop into pancreatic cancer, like an Intraductal Papillary Mucinous Neoplasm (IPMN), can take over a decade to become malignant.138-140 Thus, when PancraGEN categorizes a specimen as “Benign” or “Statistically Indolent” with a “97% probability of benign disease over the next 3 years”, the results may provide patients with a false sense of security and/or delay instituting a longer-term follow-up plan, potentially resulting in patient harm. Moreover, none of the studies available for PancraGEN follow up their entire patient populations for over 10 years. In fact, the "97% probability of benign disease over the next 3 years” is based on a 492 patient, 2015 study from Al-Haddad and colleagues where patients were followed-up from 23 months to 7 years and 8 months. Notably, 54% of the patients were followed up less than 3 years.

Third, current society and expert guidelines do not endorse or mention the current version of PancraGEN as necessary in the work-up of pancreatic cysts73,80,140,141 which further demonstrates the lack of evidence for PancraGEN’s clinical utility.

In summary, the body of literature for PancraGEN is insufficient to establish both clinical validity and clinical utility.