51˶

92-Gene Assay: Promising for Molecular NET Classification

— Identifies seven NET subtypes for those with uncertain diagnoses

MedpageToday

PHILADELPHIA -- Seven subtypes of neuroendocrine tumors (NET) were successfully classified using the 92-gene assay, researchers said here.

In a of 1,551 cases, the 92-gene assay determined small/large cell lung carcinoma was the most common NET subtype in the cohort, comprising of 50% of the diagnoses (n=749), reported Andrew Hendifar, MD, of Cedars Sinai Medical Center in Los Angeles, and colleagues at the North American Neuroendocrine Tumor Society annual symposium.

Action Points

  • Note that this study was published as an abstract and presented at a conference. These data and conclusions should be considered to be preliminary until published in a peer-reviewed journal.

The 92-gene assay also classified the following NET subtypes:

  • GI carcinoid: 14% (n=206)
  • Pancreatic islet carcinoma: 14% (n=208)
  • Merkel cell carcinoma: 10% (n=150)
  • Lung carcinoid: 9% (n=137)
  • Thyroid medullary carcinoma: 2% (n=33)
  • Pheochromocytoma: 1% (n=17)

"Neuroendocrine tumors have increased in incidence over the last 30 years, and there have been recent advancements in targeted agents and site-directed treatments that are specific for neuroendocrine tumor subtypes," explained co-author Catherine Schnabel, PhD, of bioTheranostics in San Diego, California, to 51˶. "When clinical features are equivocal in nature, molecular correlates such as the 92-gene assay [CancerTYPE ID] can address the unmet need to provide a more definitive diagnosis to select optimal treatments."

"The 92-gene assay performs well in these tumors, with 99% accuracy for molecular diagnosis of neuroendocrine tumors and 95% accuracy for tumor subtyping. We performed the current study to further define the clinical utility of the 92-gene assay ... [and] to better understand how the 92-gene assay has been integrated into clinical practice, and to characterize its impact in patients with an uncertain or unknown diagnosis."

The classifier was created based upon a database of over 24,484 tumor cases, which included de-identified patient information, as well as 92-gene assay test results. This clinical tool is able to identify the molecular classified of 28 main tumor types, as well as 50 tumor subtypes. Within this database, 6.3% of cases were successfully identified as one of the seven NET subtypes, aforementioned, which were paired with biopsy site data, patient age, and gender data.

The most common site of biopsy included liver (40%), followed by "other site," which included adrenals or kidneys (14%) and lymph nodes (13%). Among cases in which liver biopsies were performed, each of the seven NET subtypes, including thyroid medullary carcinoma and pheochromocytoma, were identified with the 92-gene assay (n=two for both).

Over 85% of gastrointestinal and lung carcinoid NET subtypes were identified from biopsy sties other than the abdomen and lungs, suggesting the tumors were potentially metastatic. Similarly, the researchers found that small/large cell lung carcinomas were the most common subtype that were identified from possibly metastatic biopsy areas, including liver, lymph nodes, and bone.

Age was a significant factor in regards to the distribution of NET subtypes, increasing with age (P<0.0001).

  • <40 years of age: n=65 cases
  • 40 to ≤65 years: n=667 cases
  • >65 years: n=758 cases

Specifically, small/large cell lung NET diagnoses were most strongly correlated with age, increasing from 25% of diagnoses among the <40 year group, versus 45% of cases in the 40 to ≤65 age group, up to 55% of diagnosis in the >65 age group. Among patients ages ≥40, this subtype was the most common molecular diagnosis.

In patients less than age <40, pancreatic islet cell carcinoma was the most prevalent subtype (37% of diagnoses), and decreased with age (P<0.0001). This subtype only comprised of 16% of cases in the 40-65 age group, as well as 10% of cases in those >65.

Gender also played a significant role in distribution of NET subtypes, with men comprising of a larger proportion of small/large cell lung carcinomas compared with women (53% versus 46%, P<0.0001). However, Merkel cell carcinoma was more common in women compared to men (12% versus 8%).

Schnabel highlighted how this classification tool can be useful in clinical practice. "[A] substantial proportion of patients were identified with a neuroendocrine tumor subtype that has an approved molecular targeted or immunotherapy drug. The molecular resolution of NET subtype rendered by the 92-gene assay can provide additional information for accurate disease staging in metastatic patients."

"While this study was descriptive in nature, future studies are aimed to characterize the clinical utility of the 92-gene assay in directing treatments by tumor subtype and results on patient outcomes," she added.

  • author['full_name']

    Kristen Monaco is a senior staff writer, focusing on endocrinology, psychiatry, and nephrology news. Based out of the New York City office, she’s worked at the company since 2015.

Disclosures

The study was supported by bioTheranostics.

Primary Source

North American Neuroendocrine Tumor Society

Hendifar A, et al "Molecular Classi cation of Neuroendocrine Tumors: Clinical Experience with the 92-gene Assay in >24,000 Cases" NANETS 2017; Abstract B-22.